CHAPTER 1

Supply Chain
Information Systems
The ab ility co access global production and services has revolutionized
business. Supply chain necworks move invenrories of various kinds from
source co consumption. Being able co work with prod11cers aro und che
wo rld provides opportunities co balance low cosr with risk mediation.
While logistics usually is associated wi1h moving material, supply chains
today can include intangibles such as services as well as invenwries o f
goods. Using the Internee enables linking cogether supply chain ne1works
in practically any business application, production o r services.
Organizations such as Dell and Hewlen-Packard have operated col-
laborative supply chains w ich each partner focusing on a few key strategic
activities. Supply chains also include organizations such as the military
and nonp rofit organizations like the Red Cross and Red Crescent. ln rhe
retail arena, Walman has been very successful in che p.1sr in linking thou-
sands of sources wich their m illions of cuscomers. Organizatio ns such
as Bank of America have viewed cheir service operarions as key ro their
success and eval uate their entire service supply chain seeking co apply
the sa me general principles as lean m an u faccu ring, focusing on provid-
ing maximum value at minimum overall cosc. Informario n systems are
needed to make these supply chains work.

Supply Chain Management

Supply chain management became a common cerm in the 1980s, heavily
infl11enced by Japa nese manufacturing processes such as those developed
by Toyora, such as just-in-time UrT) and lean manufacturing. In the
1990s electronic data interchange (EDI) made it possible to coordinate
d1.1ins of organizations worldwide. This enabled integrarion of parcicipanr
2 SUPPLY CHAIN INFORMATION TECHNOLOGY SUPPLY CHAIN INFORMATION SYSTEMS 3

supply chain elements into cooperative components sharing informa- might have literally millions of temporary sourcing arrangements. A stable
tion and enabling coordinated planning, operations, and monitoring of supply chain will have relationships benefiting all parties. Outsourcing refers
performance. There was a focus on core competencies, abandoning the to procuring sources outside the OEM organization. Outsourcing is broader,
vertical integration of Standard Oil, U.S. Steel, and Alcoa and replacing however, in that it can refer to obtaining any part of a tangible product or
it with linkages of independent organizations specializing in what they intangible service. Information systems can use EDI and web links to com-
did best. This encompassed the entire product process to include design, municate rapidly, enabling effective cost and risk management. Procurement
manufacture, distribution, marketing, selling, and service. Agile supply generally involves obtaining materials and components. Outsourcing enables
chains such as Motorola and Panasonic are flexible, enabling changing many opportunities to develop a more cost-efficient (or lower risk) supply
the set of partners for given markets, regions, or channels, accessing the chain. This comes at the cost of requiring significantly more coordination.
specific price/quality mix that enable organizations to be competitive. A manufacturing process can be developed based on what the OEM
Original equipment manufacturers (OEMs) shifted from making organization selects as the best combination of cost and risk over the total
products to become brand owners. These brand owners needed to know product life cycle. Manufacturing processes should be flexible to respond
what was going on across their entire supply chain, with the need to con- to changes in market conditions. The activities of planning, schedul-
trol from above rather than from within. Standard Oil in 1900 desired ing, inventory, transportation, and coordination across the supply chain
to control everything from within, seeking to own all elements in their require software coordination.
supply chain. Conversely, Nike doesn't make shoes anymore. They coor- Physical distribution involves moving products (or services) through
dinate activities from design to retail through communication supported the supply chain, ultimately reaching customers. The specific routing is
by a variety of information systems linked across their supply chain. referred to as a channel in marketing and can include a variety of transpor-
tation media to move goods. In a service context, the channel can involve
the routing of who a customer interacts with to get the service desired.
Supply Chain Processes
CRM is the management of the relationships between the providing
Collaboration across supply chains requires integration of all supply organization and its customers. Customer service provides information
chain activities. This requires a continuous flow of information. Key sup- from the customers and has the ability to give customers real-time infor-
ply chain processes include the following: mation on product availability, price, and delivery.
• Product development Linking independent elements to work together to deliver goods and/
• Procurement to include outsourcing/partnerships or services is flexible and enables rapid change to comply with new cir-
• Manufacturing cumstances that are commonly encountered in contemporary business. By
• Physical distribution expanding beyond the core organization, a need to monitor performance
• Customer relationship management (CRM) is needed. Some of the key measures of effective supply chain manage-
• Performance measurement ment include cost, service, productivity, use of assets, and quality. This is
often implemented through monitoring customer perceptions and identi-
Product development can be obtained by linking customers and suppli- fying best practices as benchmarks to evaluate supply chain performance.
ers. Customers can express their needs (desires), while the supply chain
organization can contribute what is possible. Communication enables
identification of a product with a competitive life cycle.
Supply Chain Information Systems
Procurement (sourcing) involves selection of supply chain members. This Many software applications are available for each step in the supply chain
can be for specific products or services, so that an organization like Walman process. Many vendors specialize in particular steps supporting part of any
SUPPLY CHAIN INFORMATION TECI INOLOGY SUPPLY CHAIN INFORMATION SYSTEMS

one of the six elements given e-drlier. Each supply chain organization will planning (ERP) systems seek co integrate all organizational information
find that they are best served by various combinations of these software systems, although of course companies will always have special needs ouc-
products. Furthermore, as technology evolves, new software is developed side of an ERP. Nonetheless, ERP systems support much of supply chain
to serve specific needs as information systems continue co evolve. activity, co include financial transactions wich sources and customers,
A supply chain management scream can be divided inco three main inventory dealings with sources, forecasting co support planning, M RP
streams: produce, information, and fina nces. co support assembly operations, and many other activities. T he crend is
for many functions chat used co be outside the ERP lo be offered as mod-
• Product-Goods movi ng from sources through manufacruring ules within ERP. One case in point is advanced planning syscem software
p rocesses and ukimately on ro a customer, ro include services (APS). There also have been systems marketed as warehouse managemenr
such as customer returns. systems, transportation management systems, manufacturing execution
• Information-Transmitting orders and updating delivery scams. sysrems, and the more general logistics management systems, targeted for
• Finances-Credit terms, payment schedu les, shipmcm, and specific industries such as the military and/or consrruction. The 2 1st ccn-
contracrual relationships. rury has seen a continued expansion of ERP systems co include add irional
functionality, such as cuscomer relationship man:igemenr (CRM) and
Because of advances in manufacturing and distribution systems, che cost SCM systems as part of the enterprise information system (EIS). There
of developing new produces and services is dropping and time to marker is also are ocher uses of information technology available to support supply
decreasing. This has resulted in increasing demand, local and global com- chains, such as online markerplaces.
pccirion, and srrain on supply chains. Supply chain management (SCM)
software links suppliers co databases that show forecasts, currcnc inventory,
Materials Requirements Planning (MRP)
shipping, or logistics time frames within the customer organization. By giv-
ing chis access to suppliers, chey can better meet chcir customers' demands. The cerm "MRP" is used as a general term co include all MRP versions,
For example, the suppliers can adjust shipping co make cerrain rhac their namely, MRP-1 (i.e., materials requircmencs planning), Closed-loop MRP
customers have the inventory necessary co meet their customers' needs. (i.e., MRP-1 with capacity planning and shop Aoor management), and
They also can monitor unexpected supply chain disruptions to organize MRP-11 (i.e., C losed-loop MRP integrated with the ocher functions such as
alternative routing. Suppliers can download forecasts into their own manu- finance and marketing).' The concept of an inccgrated information system
faccuring systems co automate their incernal processes as well. rook shape on che factory floor. Manufacturing software developed dur-
Planning applications and execution applications arc the two primary ing the 1960s and 1970s, evolving from simple inventory tracking systems
types of SCM software: to materials rcquiremcncs planning (MRP) software. MRP at its core is a
cimc-phascd order release system that schedules and releases manufacturing
• Planning applications arc capable of generating improved plans work orders and purchase orders, so chat subassemblies and components
ch rough use of mathematical algorithms arc available at the assembly station when they are required. Some of the
• Execution applications enable tracing goods, managing materi- benefits of MRP are reduction of inventories, improved customer service,
als, and exchanging financial information and enhanced efficiency and effectiveness. MRP software allows a plane
manager to plan production and raw materials requirements by working
A number of supply chain systems have evolved over the decades. The first backward from the sales forecast, the prediction of future sales. Thus, the
was materials requirements planning (MRP). This was extended co include manager first looks at marketing and sales forecasts of demand (what che
planning schedules (ofcen labeled MRP-11). Enterprise rcquiremcnrs cuscomer wanes}, che production schedule needed co meet chat demand,
6 SUPPLY CHAIN INFORMATION TECHNOLOGY SUPPLY CHAIN INFORMATION SYSTEMS 7

calculates the raw materials needed to meet production, and projects raw including chemical facilities and even universities. MRP-11 functions are
materials purchase orders to suppliers. For a company with many products, covered by production planning and ocher ERP modules.
raw materials, and shared production resources, this kind of projection was
impossible without a computer co keep crack of various inputs.
Advanced Planning Systems (APSs)
Electronic data interchange (EDI), the direct computer-to-computer
exchange of standard business documents, allowed companies to handle the Computer technology makes it possible for improvements at both the
purchasing process electronically, avoiding the cost and delays resulting from cost and value ends of the supply chain. Demand uncertainties can
paper purchase order and invoice systems. SCM began with the sharing of be better managed through improved inventory demand forecasting,
long-range production schedules between manufacturers and their suppliers. reduction of inventories, and improved transportation costs through
The MRP system should provide four basic items of information: when optimization of coordinated activities across the supply chain. Advanced
to place the order, how much to order, who to order from, and when the planning systems (APSs) provide decision support by using operational
items need to be on hand. MRP systems are used to acquire or fabricate data to analyze material flows throughout the supply chain. This sup-
component quantities on time both for internal purposes and for sales and ports the business functions of purchasing, production, and distribution
distribution. MRP is a planning instrument geared exclusively co assembly and through the entire spectrum of planning. Purchasing is supported
operations. Each manufacturing unit informs its suppliers what parts it needs by planning and MRP. Production is supported by strategic, master, and
and when it requires chem. The main aim for evolution of MRP was to cackle production planning as well as short-term scheduling. Distribution is
the problem of "dependent demand," chat is, determining how many of a supported by distribution planning and transportation planning. These
particular component is required knowing the number of finished products. planning systems interact, enabling management of demand across the
The next stage of MRP-II evolution was just-in-time OIT) method- supply chain. APS products are shown in Table 1.1.
ology in che lace 1980s. MRP-II (manufacturing resource planning) is a
method co plan all resources for a manufacturer. A variety of business func- Table 1.1. APS Software Suppliers-'
tions are tied into MRP-II systems, including order processing as in MRP, i2 Consulting and managed services
business planning, sales and operations planning, production planning, Manugistics Resource planning and supply chain management
master production scheduling, capacity requirements planning, and capac- Wassermann Smart manufacturing, distrihution and after-sales services, procurement
management, and information technology
ity planning. MRP-II systems are integrated with accounting and finance
Visopt Supply chain optimization for process industries
subsystems to produce reports including business plans, shipping budgets,
Logic-tools Supply chain optimization tools to complement existing software
inventory projections, and purchase plans. A major purpose ofMRP-II is to
Fygir Optimized scheduling and planning
integrate primary functions (i.e., production, marketing, and finance) and
Quintig Suite of supply chain planning tools
other functions such as personnel, engineering, and purchasing into the
Seehurger Business integration platform
planning process co improve the efficiency of the manufacturing enterprise.
Infor ERP with APS module option
Many within the operations management field consider ERP as a
proAlpha Multiresource, real-time supply chain optimization
natural extension of MRP-11. The APICS Association for Operations
Axxom Mohile supply chain optimization and planning
Management definition for ERP is a method for effective planning and
Epicor Acquired Vista and Vantage-ERP with APS module option
control of all resources needed co take, make, ship, and account for cus-
Oracle 11 i advanced planning and scheduling system
tomer orders. 2 There is at lease some truth to chis view, but ERP systems
SAP SAP APO
are even more comprehensive than simply on manufacturing operations.
ERP systems are found in practically all types of large organizations, So11rct". Extracted from company websites.
8 SUPPLY CHAIN INFORMATION TECHNOLOGY SUPPLY CHAIN INFORMATION SYSTEMS 9

Advanced planning systems use historical demand data as the basis of • Scheduling
forecasts that are used to manage future demand. However, in order to • Process management
optimize systems, a certain level of stability is required. John D. Rock- • Document control
efeller was able to manipulate demand for petroleum products over 100 • Data collection/acquisition
years ago, obtaining the stability he needed. Demand manipulation is still • Labor management
possible in some markets today but is much more difficult. The idea of • Quality management
supply chain optimization is more difficult to implement in conditions of • Production unit dispatch
constant product innovation, highly volatile global demand, and increased • Maintenance management
product customization (such as applied by Dell and other computer ven- • Product tracking
dors allowing customers to custom design their computer systems online). • Performance analysis
This turbulent market environment makes it difficult to obtain extensive • Resource allocation and tracking
pertinent demand history. It is easy to collect data, but demand changes
too rapidly to take advantage of it for extended periods of time. An MES can interact between the organizational ERP and the shop
floor, taking production orders from the ERP and allocating machines
and labor to tasks or products. Real status from the shop floor in turn
Warehouse Management Systems (WMSs)
is passed on to the ERP to update resource availability, track products
Warehouse management systems provide the functionality of tracking and inventory, and record production. Logistics functions in the ERP
parts throughout a supply chain. Systems such as HighJump Software and include plant production scheduling, shipping, and inventory. The MES
RedPrairie Corp offer products using electronic input such as bar code will translate that to execution in the form of dispatching, detailed pro-
scanning to track materiel through the supply chain system, maintain- duction scheduling, and tracking materiel.
ing accurate information flow to parallel physical flow. Radio-frequency
identification (RFID) technology provides another form of electronic
Transportation Management Systems (TMSs)
data input to WMSs.
Transportation management systems provide software support at an
affordable level to control shipping. A variety of alternative sources are
Manufacturing Execution Systems (MESs)
available to increase visibility and generate more efficient solutions to
Manufacturing execution systems appeared in the mid-1990s, evolving as move materiel in an increasingly complex environment involving many
all other supply chain information technology. The original focus was to risks (piracy, war, regulations). Functionality provided includes transpor-
manage demand on manufacturing organizations with respect to quality, tation mode planning, optimization models, and workflow management. 4
standards, cost reduction, schedule, and ability to react to change. With TMS software can be obtained from vendors, some of whom are
time, functions have emphasized support traceability. MES functionality listed here: 5
now integrates support to most manufacturing execution processes from
release of production orders to finished goods delivery. MES also triggers • Accuship
supply chain replenishment upstream {telling sources that replenishment • GEOCOMtms
inventory is needed). These systems use a common user interface and • HighJump {acquired Pinnacle)
data system to integrate support to multiple locations or organizations • lnfor
within a supply chain. An MES offers the following functionality: • i2 Technologies
10 SUPPLY CHAIN INFORMATION TECHNOLOGY SUPPLY CHAIN INFORMATION SYSTEMS 11

• JOA (acquired Manugistics) be entrusted to automatically and rapidly take care of everything related
• Pitney Bowes to these functions.
• Oracle (acquired G-Log) Prior to 2000, ERP systems catered to very large firms, who could
• QAD (acquired Precision Software) afford the rather high costs of purchasing ERP systems. Even focusing on
• RedPrairie a selected few modules would typically cost firms $5 million and up for
software. After 2000, demand dropped, in part because firms were often
• SAP
concerned with Y2K issues prior to 2000, which motivated many ERP
• Sterling Commerce (acquired Nistevo)
system acquisitions. Demand noticeably dropped off after 2000 came
• UPS Logistics Technologies
and went. Vendors reacted in a number of ways. First, the market con-
solidated, with Oracle purchasing PeopleSoft (who had earlier acquired
This list includes the two flagship ERP vendors: Oracle and SAP. It
JD Edwards). Microsoft acquired a number of smaller ERP software
also demonstrates the volatility of the industry, showing a number of
products, consolidating them into Microsoft Dynamics, which caters co
acquisitions. Other means ofTMS acquisition include in-house develop-
a smaller-priced market, thus serving a needed gap in ERP coverage for
ment, hosting by an ASP, or software as a service. Firms also have options
small businesses. Notably, SAP advertises chat they can serve small busi-
with respect to software within specific branches of the organization, or
ness coo. Bue it appears that they are more valuable in the large-scale
enterprise-wide support.
enterprise market. Additionally, there are many other systems to include
open-sourced ERP systems (at least for acquisition) like Compiere in
ERP France. Many countries, such as China, India, and others have thriv-
ing markets for ERP systems designed specifically for local conditions,
In the early 1970s, business computing relied on centralized mainframe
although SAP and Oracle have customers all over the globe.
computer systems. Today it is reported chat 80% of Fortune 500 firms
An enterprise information system {EIS) is appearing as a term for the
use ERP systems to manage operations.6 These systems proved their value
addition of what used to be independent add-on software such as supply
by providing a systematic way co measure what businesses did financially. chain management (SCM) systems and customer relationship man-
The reports these systems delivered could be used for analysis of variance agement (CRM) to the core ERP. This trend manifested itself initially
with budgets and plans, and served as a place to archive business data. when Oracle purchased Siebel Systems, the leading CRM provider. SAP
Computing provided a way to keep records much more accurately, and responded by acquiring their own CRM, and both vendors have added
on a massively larger scale than was possible through manual means. But SCM functionality within their systems as well. The difference between
from our perspective at the beginning of the 21st century, chat level of ERP and EIS is primarily marketing semantics, so we will use ERP for
computer support was primitive. both older and newer versions. One trend among ERP vendors is co
Business computing systems were initially applied co those func- expand their functionality to provide services formerly supplied by supply
tions that were easiest to automate and chat called for the greatest levels chain vendors such as Manugiscics and i2 Technologies.7 SAP has intro-
of consistency and accuracy. Payroll and accounting functions were an duced mySAP.com, which is an open collaborative system integrating
obvious initial application. Computers can be programmed to gener- SAP and non-SAP software. SAP APO supports supply chain activities
ate accurate paychecks, considering tax and overtime regulations of any such as forecasting, scheduling, and other logistics-related activities. Peo-
degree of complexity. They also can implement accounting systems for pleSoft has Enterprise Performance Management co support decisions at
tax, cost, and other purposes because these functional applications tend many levels. JD Edwards produces have support for planning and execu-
co have precise rules chat cover almost every case, so chat computers can tion. Oracle's 11 i advanced planning and scheduling system was designed
12 SUPPLY CHAIN INFORMATION TECHNOLOGY SUPPLY CHAtN INFORMATION SYSTEMS 13

to automate cu~tomer, supplier, and firm interactions. Vendors are mov- system. ERP systems are thus less flexible. But the benefits of integration
ing toward greater integration of supply chain products. are usually much greater than the costs ~f conformity.
The ERP concept is not applied merely for the manufacturing envi- Data can be entered once, at the m~t accurate source, so that all users
ronment but for all kinds of enterprises. Early ERP systems focused on share the same data. This can be very beneficial, because shared data is used
manufacturing, although they quickly expanded to support all sorts of more and by more people, which leads t~ much more complete and accu-
organizations. ERP facilitates enterprise-wide integrated information sys- rate data. As errors are encountered, usets demand corrections, but this is
tems covering all functional areas and performs core corporate activities limited because a set of procedures is neetled to ensure that changes do not
and enlarges customer service. ERP is a business management system that introduce new errors. This makes it harder to make corrections, but again,
seeks to combine all aspects of the organization. It is capable of taking care this added inconvenience is usually well worth the gains of data integration.
of planning, manufacturing, sales, and marketing. The concept is to inte- ERP systems also can provide bettel' ways of doing things. This idea
grate legacy systems within a coordinated integrated system. Typically, an is the essence of best practices, a key SAl> system component. The down-
ERP system uses database systems, which are integrated with each other. side to best practices is that they take a great deal of effort in identifying
Common ERP Features: An ERP system is not merely the integration of the best way to proceed with specific business functions, and that they
diverse enterprise processes mentioned earlier but also can possess key char- often can involve significant change in how organizational members do
acteristics to meet the requirements. Features often found in an ERP include their work. Further, as with any theory, what is considered best by one is
the following: often not considered best by all.
ERP systems are usually adopted whh the expectation that they are
• Best business practices-Incorporation of processes evaluated as going to yield lower computing costs in the long run. Ideally, adopting
the best in the world one common way of doing things is simpler and involves less effort to
• Comprehemive-lntegrating as many business computing func- provide computing support to an orgatiization. In practice, savings are
tions as possible, with a single database often not realized, due to failure to antitipate all the detailed nuances of
• Modular-An open system architecture allowing incorporation user needs, as well as the inevitable changes in the business environment
of those modules needed for the organization that call for different best practices and computer system relationships.
• Flexible-Capable of response to changing enterprise needs, to Training needs are typically underbud~eted in ERP projects. Further-
include Open DataBase Connectivity (ODBC) more, these training budgets don't usually include the hidden costs of
• External linkage-Capable oflinking external organizations, lost productivity as employees cope with complex new systems. Table 1.2
especially within supply chains recaps these pros and cons of ERP systems.

Among the many reasons to adopt an ERP, they offer an integrated Table 1.2. ERP Pros and Cons 8
system shared by all users rather than a diverse set of computer appli- Fact&r Pro Con
cations, which rarely can communicate with each other, and with each System integration Improved understand- Less flexibility
ing across users
having its own set of data and files. ERP provides a means to coordinate
Data integration Greater accuracy Harder to make corrections
information system assets and information flows across the organization.
Best practices More efficient methods ltnposition of how people do their work
The main benefit is the elimination of suborganizational silos that focus Less freedom and creativity
on their own problems rather than serving the interests of the overall Changing needs
Cost of computing More efficient system
organization. On the downside, ERP systems impose one procedure for planned Underbudgeted training expense
the entire organization, which requires everyone to conform to the new liidden costs of implementation
14 SUPPLY CHAIN INFORMATION TECHNOLOGY SUPPLY CHAIN INFORMATION SYSTEMS 15

The key rationales for implementing ERP systems are the following: Initially, fear of Y2K was a major concern. The Swedish survey was
later than the one in the United States, and that might explain the lower
• Technology-More powerful, integrated computer systems with
rating for this item in the Swedish study. The later Korean study did not
greater flexibility and lower IT cost
ask about this dated issue. The U.S. response was actually neutral (only
• Business practices-Implementation of better ways of accomplishing
slightly higher than 3), but Y2K clearly was a factor in ERP adoption in
tasks yielding better operational quality and greater productivity
the mid- to late- l 990s. However, more important reasons were always
• Strategy-Cost advantages can be gained through more
present. In the first two studies, replacing legacy systems received a high
efficient systems leading to improved decision making, more
positive response. The desire to simplify and standardize systems had the
business growth, and better external linkages
second highest rating in the first two studies and had the highest rating in
• Competitive advantage-If an organization's competitors adopt the later Korean study.
ERP and gain cost efficiencies as well as serve customers better,
There were two other reasons that received relatively high ratings
organizations will be left with declining clientele, competitive
in the United States (a bit lower in Sweden). These were to improve
advantage will also arise from providing better customer service
interactions with suppliers and customers, which is one way to gain
The motivations for ERP adoption were examined by three studies strategic advantage. The supply chain aspects of ERP have led vendors
using the same format. Mabert et al. (2000) surveyed over 400 Midwest- to modify their products to be more open, although work continues
ern U.S. manufacturing organizations about ERP adoption. Olhager to be needed in this direction (and seems to be proceeding). Link-
and Selldin (2003) replicated that study with 190 manufacturing firms ing to global activities was slightly positive in the U.S. survey, more
in Sweden. Katerattanakul et al. (2006) again replicated the survey, this negative in the Swedish study, and relatively higher in the Korean
study.
time in Korea. These studies reported the following ratings with respect
to motivation for implementing ERP (see Table 1.3). Three other potential reasons received low ratings in both studies.
Pressure to keep up with competitors received neutral support in the U.S.
Table 1.3. Reasons for Implementing ERP9 study. Ease of upgrading systems is a technical reason that received neu-
Rea$0ri
'
·. United States Sweden/ Korea ' tral support both in the United States and in Sweden. Restructuring the
Replace legacy systems 4.06 4.11 3.42 organization was rated lower.
Simplify and standardize 3.85 3.67 3.88 From these studies, we infer that ERP systems are an important means
systems to upgrade the quality of information systems. They can provide orga-
Improve interactions with 3.55 3.16 . 3.45 nizations with coordinated systems that have higher-quality data. Once
suppliers and customers
the kinks are worked out, this information may be available in a more
Gain strategic advantage 3.46 3.18 3.63
responsive way. Not all evidence indicates lower costs, but most evidence
Link to global activities 3.17 2.85 3.54
does indicate higher-quality information systems.
Solve the Y2K problem 3.08 2.48 NA
ERP and SCM: Originally ERP tools were not considered for
Pressure to keep up with 2.99 2.48 2.94
competitors SCM and thus the information flow between various members of
Ease of upgrading systems 2.91 2.96 3.55 the supply chain was slow. This was because until the late 1990s
Restructure organization 2.58 2.70 3.33 the concentration of organizations was on improving the internal
efficiency alone. Organizations however, soon realized that although
Racing scale from I (not important) co 5 (very important).
Source: Extracted from Maberc et al. (2000). Olhager and Selldin (2003), Kaceractanakul et al. internal efficiency is important, its benefit would be limited unless
(2006). complemented by increased efficiency across the supply chain. They
16 SUPPLY CHAIN INFORMATION TECHNOLOGY SUPPLY CHAIN INFORMATION SYSTEMS 17

also realized that, accurate flow of real-time information across the in stock or not. This is made possible because the customer's request
supply chain was the key to success in the emerging business climate, directly accesses the ERP system of the supplier.
which was characterized by rapid advances in technology, shorter
product life cycles, and so forth. Therefore, organizations started Online Marketplaces
integrating ERP applications with SCM software. This ensures that
efficiency was achieved across the supply chain, including a seam- Online marketplaces (OLMs) are exchange mechanisms that can benefit
less flow of information. ERP became a vital link in the integrated suppliers and purchasers by providing a more competitive environment
supply chain as it serves as the integrated planning and control with broader access. £-marketplaces aggregate buyers, sellers, content,
and business services. They also provide a single point of integration for
system.
interaction of buyers and sellers. A buyer can log on to an e-marketplace,
In summary, ERP applications help in effectively delivering SCM in
issue a request for proposal, and be flooded with bids. This creates a prob-
the following ways:
lem of bid comparison and interpretation. OLMs also provide services to
help sift through large numbers of bids. Different types of OLMs are
• Data sharing-They can create opportunities to share data
given in Table 1.4, demonstrating different transaction methods.
across supply chain members, which can help managers in
Auction-based OLMs are commonly used. One use of auction-
making better decisions. They also make available wider scope
based OLMs is as an exchange, seeking simultaneous bids and offers to
to supply chain managers by providing access to much broader
determine efficient prices. Future contract variants allow buyers to lock
information.
in supplies or hedge prices. Pure auctions seek only bids to establish
• Real-time information-ERP systems can provide real-time
prices for unique products. Reverse auctions do the same, only from the
information, which can be great help in supply chain decisions.
perspective of offers rather than bids. Metacatalogs focus on reducing
For example, ordering raw materials can be based on the inven-
search costs rather than on pricing. Mall-based OLMs allow buyers to
tory details provided by the ERP systems. surf a single site, with visits to individual areas representing different
suppliers.
Web-based technologies have revolutionized the way business is
conducted, and supply chain management and ERP are no excep-
tions. In order to leverage the benefits offered by this new technology Table 1.4. Types of Online Marketplace Methods
·.· ....
enabler, ERP systems are being "web-enabled." The Internet allows Transaction method
'
linking websites to back-end systems like ERP and providing con- Auction-based Exchange seeking simultaneous bids
nections to host of external parties. The benefits of such a system are Future contract variants For risk reduction
that customers have direct access to the supplier's ERP system and that Pure auction systems To establish prices for buyers

the vendors in turn can provide real-time information about inven- Reverse auctions To establish prices for sellers

tory, pricing, order, and shipping status. The Internet thus provides Metacatalogs Reduce search costs

an interface between the ERP system and the supply chain members Mall-based Access multiple suppliers at single site

allowing real-time flow of reliable and consistent information. To illus-

trate a benefit of web-enabling ERP, such a facility allows customers
to go online and configure their own products and get price informa-
tion and immediately gets to know whether the configured product is
18 SUPPLY CHAIN INFORMATION TECHNOLOGY SUPPLY CHAIN INFORMATION SYSTEMS 19

of
Example Application fl., Supply The system also kept' information on orders, receipts, and inventory
Chain Management System balances cut"rent and available in real time. When the planning module
generat~ production needs, vendor selections are automatically made
Investment in information technology has ~om~ the single large$t
considering lead times to ensure uttintertupted production. Pricing,
capital expense for U.S. firms, accounting for up to 50% of aU capital
availability checks, prodl!!=t oomparisons, and vendor selection are
investment.• Zong Dai presented a case study of the implementation Of·· instantly available. Order8 are immediately printed and distributed by
an ERP system at the corporate level with the intent ofdemonsm.ting the system, with elect~nic verification of invoices to ensure automatic
the relarion~hip of business processes with ElP. ~d how ERP systems payment to vencl0~s W;ben goods are received.
can provide sustainable competitive advantage in supply chain opera- TJ}e greatest chdenge encountered was process and change
tions. Wyeth, a global pharmaceutical research and manufacturing mana~ent; Training Wyeth personnel was challenging but was suc-
company, was studied through site visits, interviews, and examination cessfully a<:eomplisb«l through multiple methods. These methods
of documents such as annual reports, archived reconu, and electronic • included instruetor lecm~, web tutorials, and other training media.
databases. Wyeth develops and m:irkets pharmace~ieals; vaccines, and Wyeth was able to. utilize their SAP system ·to consolidate busi-
biotechnology products for both human and a.:nirnal health care. ness processes across its supply chain. The system was credited with
In 1994 Wyeth divested a number of assets that were not related improving Wyeth fin~ncial performance.
to their chosen strategi<: focus on human and animal health ~· At
that time Wyeth's proc#s were accomplkhed through a paper-driven $ourtt'; Adapted from Zang Dai (2008).
process. Thq acquired'f'ompanies that contributed to this selected
strategy. As part of their new strategy, Wy~ll ~ted to develop Conclusion
effective information technology. Their business initiatives were to
In the past, vertical integration was a way to gain efficiency in supply
transform business p~· and systems into· a globally integrated
chains. Today, vertical integration doesn't work as well, because specialty
supply chain, restruct~ri.ng produ<:t linato re<luee indirect goods and
services and improve system interoperability. An SAP ERP system organizations have developed to perform specific tasks very efficiently.
was selected with the intent of implementation by 2000 as a means Efficiency is gained today through supply chains linking specialists
to upgrade Wyeth's MRP system to implement 4ata warehousing•. A throughout the vertical business hierarchy.
component of the system w;is SAP's Advanced Planner and Optimizer A number of software systems are available to support supply chains.
(APO), software applicatiQns to support supply chain management This chapter reviewed MRP, APS, and ERP. OLM software was briefly
activities through improving production planning, prici~, schedul- described as an example of other software support. ERP systems were
ing. and product shipping. initially focused on integrating internal operations. Their high invest-
The system implementation was acoomplished by. an ·internal team ment cost and often rigid procedures made them barriers to effective
of three Wyeth IT professionals working with three SAP consultants. supply chain linkage. However, recent trends show movement toward
The greatest challenges proved to be business process reengineering. more open systems that allow closer coordination across supply chains.
The entire process from sourcing to recei~ofpayment from cusromers One way to accomplish this efficiency would be through all elements in
was to be electronically automated and srored in one central database.
a supply chain adopting the same ERP vendor products, as well as soft-
The purchasing process was streamlined through business process
ware enhancements. However, this is not economically viable for most
reengineering (BPR) to enable ~ most cost-effective procurement.
supply chain components. Many suppliers may not have the millions
20 SUPPLY CHAIN INFORMATION TECHNOLOGY SUPPLY CHAIN INFORMATION SYSTEMS 21

necessary to invest in technology adopted by the core company in the modules, organizations can cut initial implementation costs significantly.
supply chain. While vendors might argue that in the long run this might be more inef-
Other approaches are toward open ERP software. Advanced Plan- fective than full implementation now, in practice information systems
ning Systems were originally developed to enhance the ability of firms projects rarely go as planned, nor do they tend to stay within originally
to deal with other organizations in their supply chain. More recently, the planned budgets. Thus, organizations reduce risk greatly by trying par-
trend among ERP vendors is to provide this functionality within their ticular modules first, often seeing how the new system is digested by the
products, especially through Internet technology. Lean manufacturing organization, before plunging to additional modules.
is another philosophy related to gaining efficiency in production opera- There also is a difference in the difficulty of implementing different
tions. While the concepts oflean manufacturing initially seem in conflict modules. Financial and accounting modules are typically installed first,
with the idea of ERP, there have been imaginative developments allowing as they involve the most structured application. This makes it easier to
ERP systems to support lean manufacturing. implement, and easier for the organization to digest. Other modules
ERP deployment, management, and evolution are significant opera- such as materials management and planning also tend to work well.
tional concerns in today's cost-conscious business climate. The performance Conversely, support to less structured environments, such as sales and
of enterprise applications designed to streamline ERP processes and opera- marketing, tend to be more problematic.
tions is dependent on the fundamental network infrastructure. Companies
should take a holistic view of their mission critical applications and net-
Outline of the Book
working environments and include best-in-class networking solutions.
Enterprises have long made flamboyant statements about getting closer This chapter introduced various information systems available to sup-
to their customers and streamlining operations. ERP, CRM, and SCM port supply chain operations. The second chapter will more completely
applications and the organizations implementing them are in part, "bring- describe the key supply chain process of MRP and describe its relation-
ing teeth" to those superior intentions. It is not a trouble-free process, ship to ERP systems. Chapter 3 will expand discussion of ERP options,
however. In reality, the highly publicized failures of these initiatives have to include APS as an available module or as a stand-alone system. Chap-
in some minds brought concern about these applications and their pos- ter 4 will discuss the relationship of business process reengineering with
sible benefits. However, more and more organizations are moving ahead these integrated systems. Chapter 5 will present a systematic selection
with these initiatives, and the successful organizations will gain from higher technique. Chapter 6 will discuss issues in implementing such systems,
margins, better customer relations, and improved back office operations. along with demonstration of project management in the supply chain
The core idea of ERP is complete integration of an organization's software context. Chapter 7 will conclude the book, discussing three
computing system. Despite obvious advantages to vendors of each adopt- issues in implementing these systems.
ing organization installing the entire suite of modules offered, however,
only about half of the implementations seem to be of this nature. It is very
common for organizations to select modules, which makes great sense
because not every organization needs every module vendors develop. In
fact, vendors seem to recognize this through their recent emphasis on
products tailored to specific industry.
Organizations may have other very important reasons to· implement
ERP products differently than the vendors' design. A very important one
is that full system implementation is very expensive. By selecting particular
CHAPTER2

Development of
ERPandSCM
Effective supply chain operations require efficient collaboration across sup-
ply chain elements (distributors, manufacturers, suppliers) through sharing
key information for coordination. This is usually accomplished through soft-
ware tools, such as advanced planning systems or linked enterprise resource
planning systems (ERPs). Walmart has been very effective in operating at a
global level by requiring sources to have SAP systems to link to their Eiy>s. 1
Dell has developed a profitable supply chain niche in the computer field, not
by making computers, but rather through using a made-to-order e-business,
relying on a global supply chain for the parts they assemble. 2 They are only
two of many supply chain organizations that have prospered through reen-
gineering operations enabling to successfully compete. Supply chains don't
have to be pri~te organi~ions. The U.S. Department of Defense uses soft-
ware to coordinate logistics for its activities. Even nonprofit organizations
like the Red Cross coordinate their supply chains with software support.
As we stated in chapter 1, supply_chain ma~ment systems began
with materials requirements planning (MRP). These systems provided a
rational w-;; for assembly ma~facturers to control- their inventories.
- This
was extended in the 1980s to what was labeled M~-11. Parallel to that,

-
SAP developed their ERP system, centered on accounting and financial
functions. SAP continues to conduct extensive research on best practices
-
for standard business functions and incorporates the knowledge gained
into their evolving ERP product. ERP was especially at!!"!ctive to ~ufac­
turing firms, and MRP inventory management and shop-floor scheduling
....-=--
and planning were early functions supported by SAP's ERP systems. In the
last decade, more effective advanced planning systems (APSs) have evolved
to enable better decision support and control of materials flows, often in
conjunction with ERP systems.
24 SUPPLY CHAIN INFORMATION TECHNOLOGY
DEVELOPMENT OF ERP AND SCM 25

ERP Modules vendors have parallel sets of modules, as demonstrated in Table 2.2. This
information was extracted from vendor websites like www.oracle.com. As
ERP systems in concept cover all computing for an organization. The idea
with any current website, content is subject to change.
is to centralize data and computation, so that data can be entered once in
Module MM covers the functions of MRP. MRP began as an inven-
a clean form and then be used by everyone in the organization (even by
tory reordering tool in operations involving dependent demand (the
supply chain partners outside the organization) with the confidence that
demand for materials that are necessary to create the final product).
the data are correct. However, in practice, ERP vendors sold their software
The capability of MRP systems evolved to support planning of all
in modules. Modules allow clients to save money by reducing the number
company resources and currently can support business planning, pro-
of components licensed, focusing on the most important functions first.
duction planning, purchasing, inventory control, shop floor control,
ERP vendors have recently focused on offering systems tailored to specific
cost management, capacity planning, and logistics management. The
clients, such as aerospace, insurance, or medical operations. Table 2.1 gives
use of MRP resulted in better inventory and raw materials control,
a list of SAP modules around 2000 (extracted from Brady et al.3). Other
reduced need for clerical support, and reduced lead times in obtaining
materials. Improved communication and better integration of planning
Table 2.1. Modules
were also gained.
SAP Description Oracle
SD Sales and distribution: records sales orders and scheduled Marketing
deliveries, customer information Sales Relative Module Use
Supply chain
MM Materials management: purchasing and raw materials Procurement Business computing systems were initially applied to those functions
inventory, work-in-process (WIP), finished goods that were easiest to automate and that called for the greatest levels of
pp Production planning: production planning and scheduling, Manufacturing
consistency and accuracy. Payroll and accounting functions were an
actual production
obvious initial application. Computers can be programmed to gener-
QM Quality management: product inspections, material certifi-
cations, quality control ate accurate paychecks, considering tax and overtime regulations of any
PM Plant maintenance: preventive maintenance, resource Service degree of complexity. They also can implement accounting systems for
management tax, cost, and other purposes because these functional applications tend
HR Human resources: recruiting, hiring, training, payroll, Human to have precise rules that cover almost every case, so that computers can
benefits resources
be entrusted to automatically and rapidly take care of everything related
FI Financial accounting: general ledger account transactions, Financials to these functions.
generates financial statements
The degree of module use was reported by Mabert et al. (2000) and
co Omtrolling: internal management, cost analysis by cost
center replicated by Olhager and Selldin (2003). 4 Mabert et al. surveyed 479 ERP
AM Asset management: fixed-asset purchase and depreciation Asset users from the American Inventory and Inventory Control Society in the
management Midwestern U.S. in the 1990s. Olhager and Selldin patterned their study
PS Project system: R&D, construction, marketing projects, Projects after Mabert et al., using 190 Swedish manufacturing firms. Table 2.2
SAP R/3 implementation
presents information extracted from that study. Those proportions over
WF Workflow: automated R/3, task-flow analysis, prompt Omtracts 90% and under 50% are italicized for emphasis.
actions
The most popular module in the United States was financial and account-
IS Industry solutions: best practices
ing, which is the most obvious application needed by an organization. The
So11rcts: Vendor websites, 2005. Swedish study indicated that materials management, production planning,
26 SUPPLY CHAIN INFORMATION TECHNOLOGY DEVELOPMENT OF ERP AND SCM 27

order entry, and purchasing modules were just as popular. Other modules, Variants in Types of ERP Systems
given at the bottom ofTable 2.2 and each with adoption rates less than 50%,
Often firms will apply the concept of best-ofbreed, mixing modules from
either arc not considered as critical or involve less specificity in best prac-
different vendors. The Mabert et al. study found that a single ERP pack-
tices. These are similar for both studies, although human resources modules
age was utilized as the vendor designed in only 40% (56% in Sweden)
were slighcly more popular in Sweden. There have been noted differences in
of the over 400 respondents to their survey. The most common strategic
the ease in which different modules are implemented. All financial modules
tend to be relatively easy to implement. Those modules relating to manu- approach in the United States (50%, as opposed to 30% in Sweden) was
facturing and human resources also have been implemented with notable to supplement a single ERP package. In fewer cases, the idea of best-of-
success. On the other hand, modules supporting less-structured activities, breed was applied (4% in both studies). As might be expected by the
such as sales and marketing, have encountered notable implementation dif- enormity of the undertaking, few of the surveyed implementations were
ficulty. Therefore, one reason to implement ERP in modules is because of entirely constructed in-house (less than 1% in the United States, 2% in
the relative need for components of the overall system. Sweden).
Another (and probably the compelling) reason is cost. Full ERP systems The idea of best-of-breed approaches is to take advantage of what
cost a reported $5 million for very small versions to over $100 million for is perceived as specific vendor relative advantage in particular areas of
very large implementations. The fewer modules implemented, the lower application. One vendor's human resource module might be used, in
the cost. Additionally, it sometimes makes sense to implement the system conjunction with another vendor's financial and accounting system, and
in bits (phased implementation) rather than try to bring the entire massive yet a third vendor's materials management modules. In 1999, Honey-
system online at one time (Big Bang implementation). Thus, rolling out an well and AlliedSignal were merged, and the best approaches of each firm's
ERP by module sometimes makes sense as well. For a number of reasons, existing ERP systems were examined, with those components judged to
ERP in practice is usually implemented by module. be superior retained in the merged firm. 5 Quite often third-party soft-
ware designed to integrate software applications from several vendors
(midd!eware) is needed. The role of middleware products is to enable
Table 2.2. Relative ERP Module Use cross-platform operating system communications. This means that
M<>dule .· Use-Midwestern U.S • Use-:.5weden software applications such as e-commerce, data warehouses, customer
Financial and accounting 91.5% 87.3% relationship management, supply chain software, and other enhance-
Materials management 89.2% 91.8% ments can be added to ERP systems. Middleware also allows connection
Production planning 88.5% 90.5% of best-of-breed modules to the ERP backbone.
Order entry 87.7')6 92.4% If a firm chooses to utilize their own methods within an ERP,
Purchasing 86.9% 93.0% Davenport gave the choice between rewriting the code internally or
Financial control 81.5% 82.3% using the existing system with interfaces. 6 Both approaches add time
Distribution/logistics 75.4% 84.8% and cost to implementation and thus would dilute the integration
Asset management 57.7% 63.3%
benefits of the ERP. The more customization made to an ERP, the
Quality management 44.6% 47.5%
less ability to communicate seamlessly within system components
Personnel/human resources 44.6% 57.6%
and across supplier and customer systems. However, the trade-off is
Maintenance 40.8% 44.3%
that much less change in employee work processes is necessary. Thus
R&D management 30.8% 34.2%
customization reduces the hidden costs of learning the new system
Sources: Based on Maben et al. (2000); Olhager and Selldin (2003). by employees (the human side), while increasing the burden on
28 SUPPLY CHAIN INFORMATION TECHNOLOGY DEVELOPMENT OF ERP AND SCM 29

the IT staff. Not customizing makes it much easier on the IT staff Table 2.3. Master Production Schedule
but makes employees work much harder to adapt their work to the Day I 2 3 4 5 6 7 8 9 IO 11 I2 13 14 15 16 I7 I8 19 20

new system. Sedans 40 30 30 30 25 25 25 25 20 20 20 15 20 15 15 10 10 10 IO IO

Another important concept is the idea of federalization. Davenport Roadsters 15 I8 17 13 15 I6 I4 14 13 13 11 9 10 9 8 7 6 6 6 5

used this term to describe the process of rolling out different versions Town 10 10 10 10 IO IO 10 10 10 IO 10 10 10 10 10 10 10 10 10 10
cars
of an ERP system in each regional unit, tailoring each location's system
SUVs 60 30 29 29 26 25 24 24 23 23 23 22 22 2I 2I 20 20 I9 18 I8
to accommodate local operating practices.7 Hewlett-Packard, Monsanto,
and Nestle have all used this approach, establishing a common core of Table 2.4. Inventory and Purchasing Infonnation
ERP modules shared by all units but allowing other modules to be oper- Item Initial on Order Lead On order
ated and controlled locally. hand conditions time
Because of its focus on a key function of supply chain management, Assembled small I50 I day
we will describe the basic MRP process in greater detail. (Specific com- engine
Assembled large I5 I day
mercial software may differ in details.)
engine
Assembled 1200 1 day
Materials Requirements Planning wheels
Assembled doors 1000 1 day
The MRP system begins with three documents. A forecast for end- Engine-small 100 Each 3 days
items being assembled is needed by time period. A bill of materials Engine-large 40 Each 4 days
(BOM) describes the components that go into an assembled product. It Chassis-Sedan 80 Each 2 days
lists each part in a hierarchical tree, by quantities required for each sub- Chassis-Roadster 50 Each 2 days 30 in day 2
assembly, all the way up to the final end-item. Finally, inventory records Chassis-Town 30 Each 3 days 20 in day 2, 10 in day 3
car
describing quantities of each component on hand is needed, as well
Chassis-SUV 80 Each 3 days 20 in day 1, 30 in day 3
as managerially determined ordering policies with vendors and lead
Battery 173 Multiples of 12 2 day
times.
Wheel 1000 200minimum 2 days
We use a small example to demonstrate the basic workings of MRP,
Windshield 250 SO minimum 2 days
one of the most important supply chain management processes. MRP
Dtx)r 1200 Multiples of 400 1 day
begins with a master production schedule showing the requirements
Hood 150 SO minimum I day
for output from the manufacturing facility by time unit (Table 2.3). A
Rim 200 Multiples of 200 2 days
second component of the system is inventory and purchasing informa-
Tire 300 Multiples of 144 1 day
tion, which shows the initial quantity on hand by item, order conditions,
Lock 112 Multiples of 1 day
lead times, and currently open orders (Table 2.4). Additional inventory 1000
information such as safety stocks could also be included. BOMs show Handle 332 576minimum 2 days
the hierarchy of components required to manufacture each end-item and Window 138 SO minimum I day
subcomponent (Figures 2.1 through 2.4).
Assume an automobile company produces four types of automobiles.
They use a rolling 20-day planning horizon.
30 SUPPLY CHAIN INFORMATION TECHNOLOGY DEVELOPMENT OF ERP AND SCM 31

Assemble Assemble
Sedan Town car
J./,ey I day

A.skmhlc As~l'mhlc
enginl:' cn~inc

t .lay 1 Jay

Chassis-- As~mhk WinJshickl Hood Cha!iSill- ~'l'rahk WinJshidJ Hlllll.I

~.lan(1) u.'h ..... ls (4) (I) (I) T.lWncar (1) ui.,,/, (4) (I) (I)
l J11y ltkry

noor{t) Luck(I) HanJlc-(1) WinJow(I) Door(t) L><:k(I) Handk· (I) WinJnw(I)

(Oun hanJ) w.mhanJ)
En~inl:' small (I) Eni:i:inc larJ.'C' (I)
llatrl:'ry(l) B;mcry(I)
(0 un hanJ) (Oon hand)

Rim(I) Wh«l(l) Tin(I) Rim(I) Wh,cl(I) Tire (I)

Figure 2.1. BOMfor sedans. Figure 2.3. BOM for town cars.

Assmtblc Assemble
Roadster SUV
I Jay ldtcy

Asse!mble ~cmhl..:
e:-ng:ine cnJ,,rinc
1 .lay I Jay

Chassis- ~mbk Win.lshiclLl H00<! ASSi:mhk

ChassisSUV(I) Windshield Hoo1..i
RuaJster (I) tfltet.'l~ (4} (I) (I)
tdll.-.:ls(4) (I) (I)
lili1:y JJi1,

Duor(I} Lxk(ll Hm1.1le(I) Win.luw(l)

lloor(l) '-'"k(I) Handle(!) Window(!)
(0 unhand)
(O on hand)
En~inesmall (I) En.i:incsmall(I)
Battery(l) Batrcry(l)
(Oon ha11l1) (0 un han.I)

Rim(l) Whed{t) Titt{l)

Rim(l) Whed(I) Tire(!)

Figure 2.2. BOM for roadsters. Figure 2.4. BOM for SUVs.
32 SUPPLY CHAIN INFORMATION TECHNOLOGY DEVELOPMENT OF ERP AND SCM 33

Vehicles are assembled with the appropriate chassis, engine, wheels, Table 2.5. Top-Level Quantities Required
windshield, and doors. Assembly of vehicles takes one day: Materials are Assemble Day Day Day Day Day Day Day Day Day Day
received ready for use at the beginning of the day (lead time includes Roadster 1 2 3 4 5 6 7 8 9 10
unpacking and preparation). Small engine 1 ea 15 18 17 13 15 16 14 14 13 13
assembly
Engines, wheels, and doors are subassemblies. Each of these subas-
Chassis- 1 ea 15 18 17 13 15 16 14 14 13 13
semblies takes one day. Roadster
Wheel 4 ea 60 72 68 52 60 64 56 56 52 52
• When an engine is received, a battery is inserted to make an assembly
assembled engine. Windshield 1 ea 15 18 17 13 15 16 14 14 13 13
• When a wheel is received, a rim and tire are added, making an Door assembly 2 ea 30 36 34 26 30 32 28 28 26 26
assembled wheel. Hood 1 ea 15 18 17 13 15 16 14 14 13 13

• When a door is received, a lock, handle, and window are

Assemble Day Day Day Day Day Day Day Day Day Day
added, making an assembled door.
Town car 1 2 3 4 5 6 7 8 9 10
Large engine 1 ea 10 10 10 10 10 10 10 10 10 10
Parts needed at the beginning of each day for assembly of each vehi- assembly
cle, shown in Table 2.5. Chassis-Town 1 ea 10 10 10 10 10 10 10 10 10 10
car
Wheel 4 ea 40 40 40 40 40 40 40 40 40 40
assembly
Windshield 1 ea 10 10 10 10 10 10 10 10 10 10
Door assembly 4ea 40 40 40 40 40 40 40 40 40 40
Hood 1 ea 10 10 10 10 10 10 10 10 10 10
Table 2.5. Top-Level Quantities Required
Assemble Day Day Day Day Day Day Day Day Day Day Assemble Day Day Day Day Day Day Day Day Day Day
Sedan 1 2 3 4 5 6 7 8 9 10 SUV 1 2 3 4 5 6 7 8 9 10
Small engine 1 ea 40 30 30 30 25 25 25 25 20 20 Small engine 1 ea 60 30 29 29 26 25 24 24 23 23
assembly assembly
Chassis-Sedan 1 ea 40 30 30 30 25 25 25 25 20 20 Chassis-SUV 1 ea 60 30 29 29 26 25 24 24 23 23
Wheel 4 ea 160 120 120 120 100 100 100 100 80 80 Wheel 4 ea 240 120 116 116 104 100 96 96 92 92
assembly assembly
Windshield 1 ea 40 30 30 30 25 25 25 25 20 20 Windshield 1 ea 60 30 29 29 26 25 24 24 23 23
Door assembly 4ea 160 120 120 120 100 100 100 100 80 80 Door assembly 4 ea 240 120 116 116 104 100 96 96 92 92
Hood 1 ea 40 30 30 30 25 25 25 25 20 20 Hood 1 ea 60 30 29 29 26 25 24 24 23 23
34 SUPPLY CHAIN INFORMATION TECHNOLOGY
DEVELOPMENT OF ERP AND SCM 35

Requirements can be aggregated across all vehicles (Table 2.6). Table 2.7. Small Engine MRP Worksheet for Day I
The MRP analysis for each of these elements can then be conducted, Assembled Order Day Day I~ay Day Day Day Day Day Day Day
for instance, for small engines (Table 2.7). small engine info 1 2 3 4 5 6 7 8 9 10
This tells the assembly operation that 43 small engines need to be Gross 115 78 76 72 66 66 63 63 56 56
assembled on Day 1. The MRP analysis is conducted each time period, Requirements

or daily, because there might be many changes in demand, receipts, or Scheduled

receipts
inventory on hand. The only action taken is for Day 1. If, as is the case
On Hand 150 35
here, items are ordered lot-for-lot (L4L in inventory code), the rest of
Net L4L 43 76 72 66 66 63
the time period data can be calculated and treated as a forecast of future 63 56 56
requirements
demands. However, as we will demonstrate, if ordering complications Planned order l day 43 76 72 66 66 63 63 56 56
arise (orders come in multiples of some value or with some minimum release lead
order}, they will become scheduled receipts in subsequent analyses. You
can't update this particular day's MRP form, because that will create
a cycle. (This little problem can be solved in many ways, but for our The schedule for assembling large engines is outlined in Table 2.8.
purposes it is sufficient to treat MRP as a time-period by time-period Plant management needs to know that five large engines need to be
calculation.) assembled on Day 1.
Knowing the planned order release for assembled small engines and
assembled large engines now allows us to proceed down the bills of
material to identify subcomponents that need to be ordered. (Assem-
Table 2.6. Aggregated Part Requirements bled small engines consist of one small engine and one battery, while

Day Day Dav Dav Dav Day Day Day Day Day assembled large engines consist of one large engine and one battery.)
1 2 3 4 5 6 7 8 9 10 The planned order release for the higher-level component becomes the
Assembled lea 115 78 76 72 66 66 63 63 56 56 gross requirement for the subcomponents (multiplied by the number
small engine each).
Assembled lea 10 10 10 10 10 10 10 10 10 10
large engine
Chassis-Sedan lea 40 30 30 30 25 25 25 25 20 20 Table 2.8. Large Engine MRP Worksheet for Day I
Chassis- lea 15 18 17 13 15 16 14 14 13 13 Assembled Order Day Day Day Day Day Day Day Day Day Day
Roadster
large engine info 1 2 3 4 5 6 7 8 9 10
Chassis-Town lea 10 10 10 10 10 10 10 10 10 10
Gross 10 10 10 10 10 10 10
car 10 10 10
requirements
Chassis-SUV 1 ea 60 30 29 29 26 25 24 24 23 23
Scheduled
Assembled 1 ea 500 352 344 328 304 304 292 292 264 264 receipts
wheels
On hand 15 5
Windshield lea 125 88 86 82 76 76 73 73 66 66
Net L4L 5 10 10 10 10 10 10 10 10
Assembled 1 ea 470 316 310 302 274 272 264 264 238 238 requirements
doors
Planned order l day 5 10 10 10
Hood lea 125 88 86 82 76 76 73 73 66 66 release lead
10 10 10 10 10 -
36 SUPPLY CHAIN INFORMATION TECHNOLOGY
DEVELOPMENT OF ERP AND SCM 37
Table 2.9. MRP Worksheets for Engines
Here the cycling complication can be demonstrated. The batteries
Engine-small Order Day Day Day Day Day Day Day Day Day Day on hand will cover demand through Day 2, and all but 13 required on
info 1 2 3 4 5 6 7 8 9 10
Day 3. A 2-day lead time means that orders must be placed in Day 1 to
Gross 43 76 72 66 66 63 63 56 56 - cover at least 13 batteries. But batteries have to be ordered by the dozen.
requirements
Scheduled
One dozen (12 batteries) would be insufficient. Therefore, two dozen (24
receipts batteries) are ordered. On the next day's MRP form, these 24 batteries
On hand 100 80 50 20 would show up as scheduled receipts on what is now day 3. At that time,
Net requirements L4L 5 25 25 25 20 20 - this will cover the demand for batteries for that day, and we can see that
Planned order 3 day 5 25 25 25 20 20 - - - 11 of these batteries will be carried forward to what is now day 4, leaving
release lead a net requirement of (76 - 11 = 65) 65 batteries, which will in turn trig-
ger an order for 72 batteries (the nearest dozen covering 65). Therefore,
Engine-large Order Day Day Day Day Day Day Day Day Day Day
info 1 2 3 4 5 6 7 8 9 10 we will only worry about planned order releases for the current day.
Gross 5 10 10 10 10 10 10 10 10
requirements
Advance Planning Systems
Scheduled
receipts Advanced planning systems (APSs) to aid supply chain material flows
On hand 40 35 25 15 5
have been developed. 8 They can be obtained independently from an ERP,
Net requirements L4L 5 10 10 10 10 - added on to an ERP, or found as modules within an ERP. Major ERP
Planned order 4 day 5* 10 10 10 10 - - - - - vendors have added APS modules. These systems offer support to meet 1:
release lead
the need to quickly and efficiently plan and control supply chains in
*The planned order release for Day 1 is the action item, identifying the need to place an order. Other dynamic environments. They consider resource capacities and can handle
day planned order releases will await more current information tomorrow.
multiple sites and transportation links, offering near-optimal plans over
the entire supply chain, and have the ability to identify bottlenecks. APSs
The number of batteries to order depends on the demand for both
may include modules to support the following:
engines. The sum of planned order releases for both small and large
engines becomes the gross requirements for batteries (Table 2.10).
• Supply network design through demand planning
Table 2.10. MRP Worksheet for Batteries • Supply chain planning to include purchasing and materials
planning
Batteries Order Day Day Day Day Day Day Day Day Day Day
info 1 2 3 4 5 6 7 8 9 10 • Demand management and forecasting
Gross 48 86 82 76 76 73 73 66 66 - • Production planning and scheduling
requirements • Distribution and transportation planning
Scheduled • Execution to include order release, shop floor control, vehicle
receipts
dispatch, and order management
On hand 173 125 69
Net requirements Xl2 13 76 76 73 73 66 66
APSs have been credited with enabling planners to better visualize what
Planned order 2 day 24
release lead
they could make given constraints for specific operations, ranging from
shop-floor conditions to warehousing constraints and transportation
38 SUPPLY CHAIN INFORMATION TECHNOLOGY DEVELOPMENT OF ERP AND SCM 39

system limits under conditions of specific demand requirements. They

have enabled much more efficient service in meeting changing customer 2..:enters

demand under conditions of rapidly changing prices. APSs enable chang-

ing plans rapidly, providing what-if analysis to change product mix.
Industry cases demonstrate conditions under which APSs have pros- Fipre .i!s. Associanon se~ supply ~fudn.
pered and when they have failed. 9 The more uncertainty in the process,
A.. La"°n M3 Supply Chain Plannin{(SCl?) system was ~p1Ci1
the greater the challenge of keeping human participation needed to make
mented in ~004. The system was. viewed as a decisiop support ~~ .
the systems work. Human interaction was found to be needed, as relying
l!he ·~tem balanced supply a1.1d demand for each weekly tifue period
on autonomous APSs has led to problems. during the ·planning horizon (the remainder Qf the planting··seasqn).
The following case demonstrates some of what an APS does and how
This is ~iiniw ~o the .MRP planO.ing h~rizOri. Input data were fi?reca.sts ·
it can enable better management of a supply chain.
(as in MRP); customer orcl~rs (realized. demands); raw material availa~le
(inventory data); and<capacities in~ll\S,()f pl'QductiQn. warehouse stqr.,
Ad~qced Planning Systems in age, and.. transportatiQn•. Forecasts were genel!lted.by qie annual budget
a~d were only updat~ twi~ per year. Actual customer orclers ~re used
Swenish Seed Dis(ributlot. .
",\<: . . "'·:,. <'. . ...;: ' . '>. for the first 2 weeks of the planning
' : .
horii.on. Capacity and raw material
.
Tht'.~~sh H\rme,{S Silpl'ly and Crop Marketing Association is a data were obtain~ froin the Association's.ERP ~tel;fi and feld into the
prod.u~~ ~perative 'proyiding .marketing, distribution, sale5, .pro- Lawson M3 SCP system. P.-oduction and inventory levels were tn,atched
,;:ces;irig;(a"1d $~ppfy to~bou,,t 30,000 farmet$ in SwCd~n. Prior to 2001. wit\i capacity fo:r each peri~. ponsidering the four production an4 two
.sw~~ lrmed:i~li<ld ~tt I~• and regional cooperatiW:s 'for seed warehouse facilities. I.inear ;µid m,i?ced~integer programming was used
su~eli~: 'r4~.\Vere.~erg~intothe As~iation fo·2ol>linto 13geo- by the M3 SCP to minimi7.C cost over the overall supply chain system,
. 'tta~i~~ ~~ ti?. p~vid~rai¥iets \Vith.seed/fertilizers, fef:d;•and other with. simulations conducteld based on the gene~teld solutions,,
r,ct<IJ~<ld.ma~erials. The;i~~ci~~ion also sells what is prod.ufed. They' A four-phase iterative. plann~ng system was u.~d.. In tbe first pltase,
: h~v~~ild3;000.~qipl6yees ~d iuar~t f~odto 19,.copn~!es. . ·.··
. <\P unConscrajned master plan w.as established to see wh~re ~urces
we~e overloaded.• Thi~ would be much the same' as tbe MRP .system
'jhe .Ass9~ati~n handles· the enti~~ seed supply C¥°:k~ginnin~
described eadier. Th~,second phase took this output and reran the sys-
\vidl cl~siqg ra\V set!d before pacJcing and diS\'fibu~io.~~ and deafS
tem, ~dding constrained production, especially for.seed cleansirui; and
>:with 270 ..it~~s (including dfffe~nt package sizes},deliv~riog about
packaging capacities CQU,sidering minimum batch siz.es to avoid expen~ . .
~\~ t<>;15 of~ a~e~ duri~;\h,e planti~g ~ri•. ·.. , • sive start-up~, This yielded a pl~n for the numl:>er of shifts te<{µired at
<A.·;qtaj()r\1'esm1ctµnng ohhe seed supply chant was condu~ed in each production facilities. The .third phase add<ld limits on wa,rehouse
2494,~: T. ';)f six p~ddctiq~ plants were closeld, ;ilong with two of capacities, and new solutions were generated. To this point, unUm.,
bfo~f tenrt.J\Varthouses. Tqi;·multedin lower production and storage iced transportation capacity was ~umed. The fourth and final phase
.cap~cltr• th.ti~· incl-easing r~uire1;t1ents on the transportation system. added transportation constraints.
. The ~a?,tpn !~pply chkin consist~ is illustrateld in Figure 2S. Use of che system streamlined· prod1.1ction and.dis(~ibution. To~
Th~~~ati9ricontrolleldproduction andstoragc facilities. Trans- costs were decreased about 13% annually despite increasing the quantity
~~t~iion ~ outS<)urced but needed assignment information with of goods sold. A more precise measure of value added was a reduc-
aboti~ 2 weeks oflead time. tion of total cost per ton by about 15%. The use of the APS resulted
40 SUPPLY CHAIN INFORMATION TECHNOLOGY

inincreasecl:pred~tiQn "tch~siles because pr~u~tion was eentraliir.ed

to four ·radter. dian(six>pta.i~.. T~nation·«>sts inc~ slightly
because more ~t ~ibut~ ~·Used. ~~ng to overall system .
CHAPTER3
efficiency. Less C4lpi¥d V{a$ ~~ed formvent@ry ~e to better system
tJiroughput. Total pianG:in1,time ~ reducetl through the APS, with Supply Chain Management
incr~~reol1trol of ma~al ftl;l)WS,, . .

So~rre. ¥apml
,, from ~\!erg~<l V,iu1',n (2009);
,. ." .. '.~:, ',·,"
Software Options
Conclusion
In the past, organizations typically developed their own software using
A variety of types of software are available to support supply chain man- their own in-house programmers and staff. This led to development of
agement. The most important functions are forecasting, planning, and many useful but diverse and unconnected legacy systems. During the past
control of materiel throughout the system. MRP is a system designed to few decades, many software vendors have focused on developing soft-
take forecasts and use them to control materiel in assembly operations. It ware for specific applications, creating many specific software products,
requires a relatively stable operating environment in order to be effective to include materials requirements planning (MRP) systems, advanced
{lots of change in actual demand versus forecasted demand causes negates planning systems (APS), and enterprise requirements planning (ERP)
many of the benefits of reduced inventory). systems. Thus the software field has evolved from each organization cre-
Advanced planning systems are focused on supply chain activities of ating its own software to a vendor system selling off-the-shelf packages.
forecasting, sourcing, and monitoring inventory. As such, they deal with An even later development has been open-source software products.
the key processes related to supply chains only. They can be obtained
stand-alone, added on to an ERP from an outside vendor, or obtained as
Options
a module within an ERP, at least from the larger ERP vendors.
ERP systems are rarely obtained with all available modules. Not every There are many approaches to provide information systems to support
organization needs every module vendors have available. Even if a client supply chains. The most well-known are SAP and Oracle. There are many
might be able to use some particular module, they may not find it cost- other commercial vendors as well. And commercial vendor software is by no
effective and can reduce their ERP licensing and maintenance costs by means the only source of software to manage supply chains. There are appli-
eliminating marginally effective modules. The module system makes it cation service providers, and organizations can develop their own systems
possible to tailor ERP systems to particular organizational needs. in-house (although as we will stress, this involves a lot ofwork). The primary
One of the most important modules for manufacturing involves MRP categories of options to obtain supply chain support are the following:
functionality. As with APS, MRP software can be obtained stand-alone,
added on to an ERP from an outside vendor, or obtained as a module • A custom system developed in-house
within an ERP. • A stand-alone advanced planning system
While having the ability to choose specific modules provides flexibil- • A full vendor ERP system
ity, it also complicates matters. Chapter 3 will discuss optional forms of • Selected vendor modules
supply chain information system support in greater detail. • A customized vendor ERP system
• A best-of-breed approach
42 SUPPLY CHAIN INFORMATION TECHNOLOGY SUPPLY CHAIN MANAGEMENT SOFTWARE OPTIONS 43

• Application service providers system budgets. There also are many hidden costs, including heavy con-
• An open-source system sulting fees, annual maintenance contracts, unexpectedly large training
costs, and complications from changing how employees do their jobs.
There are many competitive vendors to SAP and Oracle. One notable
Customized Systems Developed In-House vendor is Microsoft, offering Great Plains ERP software at a price dearly
The traditional (and very much outdated) approach is to develop software targeted to reach intermediate-sized organizations (with annual information
using only in-house assets. This approach does offer the greatest opportu- system budgets around a million dollars). Additionally, there are many other
nity to gain competitive advantage. However, creating in-house software smaller vendors, making the vendor choice for small- to intermediate-sized
on this scale is a very difficult information systems project. Developing organizations very interesting (and complicated). Generally, you pay for what
a strong advanced planning system in-house calls for hiring a great deal you get. But not everyone needs the computing power suitable for Exxon.
of expertise that cannot hope to compete with vendors specializing in a
particular software product. The ideal way to develop an integrated sys-
Selected Vendor Modules
tem is to combine it with extensive business process reengineering (which
we will discuss in chapter 4), identify the best way to do everything, and Compromise systems are available. Most firms adopt only a few modules
then build the computer system to accomplish this. This is a very slow of vendor software. This is a partial form of an ERP system, which has
and expensive way to obtain supply chain software. Yet this approach is the relative advantages of minimizing organizational risk and expenditure
the most flexible and responsive to organizational needs. in the short run and minimizing the trauma of incorporating the ERP
system into organizational operations. The disadvantage is that the full
functionality of the vendor system is not obtained, and users still must
Stand-Alone Advanced Planning Systems (APSs)
conform to the procedures that the vendor programmed into the system.
Stand-alone APS software was presented in chapter 2 and is widely avail-
able. This will be quite a bit less costly than a vendor ERP system but
Customized Vendor ERP Systems
will only provide specific planning support for supply chain operations.
It would make a great deal of sense to integrate an APS with financial, Many companies adopt another hybrid approach, customizing a vendor
accounting, and other organizational computing. This integration is the software product. This gains flexibility over simply adopting the vendor
essence of an ERP system. system but risks loss of the efficiencies built into the system through best
The simplest option (and by far the most expensive) is to adopt a full ven- practices. This approach has the advantage of retaining core company
dor ERP product. This is the option that vendors will suggest, and it has some competencies, or methods that the organization does very well, along
strong advantages, especially with respect to relative time and installation. with some of the advantages of integrated systems. There is much less
negative impact on how employees do their jobs. However, it makes the
IT staff work much harder and spend a great deal more money.
Full Vendor ERP Systems
SAP and Oracle have become world leaders in software, both offering
Best·of·Breed Approaches
very powerful systems capable of supporting multinational organizations.
But they are expensive. While SAP and Oracle contend to offer products The best-of-breed approach is similar to selecting a few vendor modules.
suitable to small business organizations, they really aren't interested in Each vendor has developed a reputation for some specific ERP function.
talking to organizations without multimillion-dollar annual information Modules considered to be competitively strong are selected from multiple
SUPPLY CHAIN MANAGEMENT SOFTWARE OPTIONS 45
44 SUPPLY CHAIN INFORMATION TECHNOLOGY

and an industry focus on SaaS, where vendors develop software posted

vendors. Dial, a manufacturer of soap and other products, at one time
to the web and made available for customer use. Web delivery has been
had a mix of applications from Siebel, Oracle, Manugistics, and other
selected as a means to distribute a number of interesting enterprise sys-
sources. 1 Using the best-of-breed approach, custom interfaces can be
tem software, led by Compiere from France. Compiere and many similar
developed using in-house information system development assets. Over-
products are not open source in the sense that users can modify the code.
all, however, it usually creates more trouble than it's worth.
They are open in the sense that they are downloadable for free. The busi-
ness model is based on collecting fees for service and support.
Application Service Providers (ASPs) The first six means of obtaining supply chain software support listed
so far are fairly traditional. We will elaborate a bit on two relatively
You can in effect rent an ERP system through an application service pro-
unique approaches: outsourcing and open source.
vider (ASP). This is a form of outsourcing. Dial reorganized its IT and
installed an SAP suite run by Electronic Data Systems (EDS) to replace
Dial's 50 IT employees, who were to be transferred to EDS after an 18- Outsourcing Supply Chain Management Software
month, $35 million project. Only a small governance team was retained
There are risks in outsourcing. In many cases, costs rise precipitously
by Dial, with the purpose of dealing with IT strategy, architecture, and
after the outsourcing firm has become committed to the relationship.
industry applications. Overall expenses for the transfer were expected to
These tradeoffs are recapitulated in Table 3.1.
be $110 million. Outsourcing is attractive to many types of organiza-
tions but especially to those that have small IT staffs, without expertise
Table 3.1. Factors for and Against Outsourcing
in enterprise systems. The primary benefit of an ASP is that the using
Reasons to. outsburce Reasons agaigst <IU~tcing
organization doesn't have to worry about system development, nor about
Reduced capital expenditure for software Security and privacy concerns
being at the mercy of vendors when they make changes to their software. and updates
Some organizations, such as General Motors, have outsourced all their IT Lower costs gained through ASP and Concern about vendor dependency and
operations. 2 However, the risk is simply transferred, because the user is economies of scale (efficiency) lock-in
now subject to the mercy of the ASP. The decision is very similar to that More flexible and agile IT capability Availability, performance, and reliability
of deciding to buy or rent housing. In the long run, you are usually better concerns
Increased service levels at reasonable cost High migration costs
off buying a house. However, the cash flow impact and risk avoidance of
Expertise availability unaffordable in- Expertise is a competency critical to orga-
renting is much better than buying. house (eliminate the need to recruit IT nizational success
personnel)
Allowing the organization to focus on Systems are inextricably tied to IT
Open·Source ERP Systems their core business. infrastructure
Another major development in ERP has been open software systems. Continuous access to the latest technology Some key applications may be in-house
The open software idea was made famous by the Linux operating system. and critical
Reduced risk of infrastructure failure Operations are currently as efficient as the
There are many nuances to definitions, but we will use the term open
ASPs
source for software distributed without charge. A related idea is software
Manage IT workload variability Corporate culture doesn't deal well with
as a service (SaaS}, involving web distribution of bits of software capable working with parrners.
of doing particular functions, at a fee. Open-source software provides the Replace obsolete systems
opportunity to utilize more service-oriented systems, strongly supported
Source: Bryson and Sullivan (2003); "ERP ou1sourcing" (2003); Clymer (2004); Olson (2004).
by organizations such as IBM with their on-demand computing initiative
46 SUPPLY CHAIN INFORMATION TECHNOLOGY SUPPLY CHAIN MANAGEMENT SOFTWARE OPTIONS 47
One risk is that an ASP might shirk its commitments due to bankruptcy information systems (EISs) as discussed in chapter 1. Recently, ERP ven-
or for other reasons. ASP sites might also be attacked and vandalized, or dors have realized that open-source systems have capabilities, both as a
destroyed by natural disaster. Each organization must balance these factors source of content for vendors and as a threat to the proprietary enter-
and make their own decisions. The case studies in this chapter demon- prise system market share from competitors based on OSS development
strate how two organizations reached different conclusions. or delivery. Open-source ERP products can provide flexibility and of
ERP is only one service offered through ASPs. Acquisition of supply course have the advantage of free software access. As ERPs are commonly
chain management systems such as an advanced planning system through implemented by organizations, it is hard to attain competitive advantage
an application service provider often makes sense. through implementation of an ERP product. OSS ERP systems can be
an answer for competitive advantages since organizations are able to cus-
Open-Source Supply Chain Software tomize their information systems by modifying the open software codes.
Three potential benefits in using OSS ERP systems are increased adapt-
Open-source development has proven highly successful in general soft- ability, decreased reliance on a single supplier, and reduced costs.
ware product development. Red Hat claims that open-source software
(OSS) can save businesses money by Comparison of Supply Chain
Management Software Sources
1. Enabling use of commodity hardware rather than proprietary machines
Table 3.2 compares advantages and disadvantages of some representative
2. Avoiding expensive maintenance contracts
points on a continuum.
3. Obtaining greater functionality, reliability, and performance
4. Increasing productivity through a faster learning curve and availabil-
ity of support tools
5. Avoiding vendor lock-in Customized systems I Best lit with organizational needs Most difficult to develop
developed in-house
6. Reducing the need for specialized security consultants and tools Most expensive
Slowest
Stand-alone APSs Less expenditure Harder to integrate with other
Open-source software is thus becoming a viable alternative to propri- Much simoler installation aoolications
Full vendor ERP Relatively fast Inflexible
etary software, with an obvious cost advantage. There are risks, in that products Less expensive than customization Make employees change work
one cannot expect the same level of service with OSS as with proprietary Efficient from an IT perspective methods
Easier to ulll!Tilde
alternatives. However, support for many OSS products is available, from Selected vendor Less risk If expanding, long run time
modules Relatively fast
such organizations as IBM and Red Hat. Contemporary software selection and higher costs
Least exoensive vendor aooroach
thus requires considering the tradeoffs between open-source and propri- Customized vendor Retain flexibility while keeping Slower
ERPsvstems vendor exoertise
etary software. Open-source ERP products include Compiere, OpenMFG, Usuallv more exoensive
Best-of-breed Theoretically gain best of all systems Difficult to link modules
Open For Business Project, Tiny ERP, Open Office, and OpenPro, each approach
Slow
providing various levels of enterprise information system functionality in Application service
Need middleware
Least risk At the mercy of ASP
various forms of open-source relationships. The open-source project center providers Least cost No control
Fastest
Sourceforge.net had over 1,000 ERP projects ongoing as of May 7, 2009. Subject to price increases
Least subiect to vendor chanl!e
ERP systems have evolved to expansion of functionality, especially Open-source systems COST! (acquisition is free) Greatest risk (other than
Flexible in-house)
in the form of customer relationship management (CRM) and supply
Need employees with ability
chain support, to a transformed product often referred to as enterprise to link ooen-source svstems
SUPPLY CHAIN MANAGEMENT SOFTWARE OPT!l)NS 49
48 SUPPLY CHAIN INFORMATION TECHNOLOGY

We will conclude this chapter with a case study of a small organi- and based on the average implemen~tion cost of $60,000, all were
zation's experience in obtaining inexpensive software support for their rejected. The owners felt that at that price it would be worth recon-
made-to-order planning operation. sidering .in•house .development. Both owners had come from larger
companies and had ~en some of the horrors of vendor ERP imple-
Implementation of an Open...Sour:c'°iRP System mentation. They felt that the high cost was not justified or worth the
headache of drastically ~n,forming to a system that would lock them
WETI (fictitious name of a real firm} is a small ent«prise that manu-
into ongoing support and main'tenance costs for the life of the soft-
factures industrial automated welding systems for several industries.
ware, as is typical of tra.ditional ERP software vendors.
As the client base expanded the required inventory to service the
many uniquely engineered to order systems expanded as well. Cus- First Round Choice: ERPlite
tomer service for repair and replacement puts also bef;:ame a large
part of daily operations. Demand for systems increased, the number The employee in chuge of finding a software solution examined low-
of engineers grew and the subseque,nt paper trail .exploded. There cost ERP systems available at that time. ERPlite. (http://www.erplite.
was a need for a system to manage orders~ scht:duling, billing, and com) was selected. The price per user was around $300 at the time,
which was drastically lower than the prices offered by the traditional
inventory.
ERP vendors. Furthermore, the software seemed that it would meet
BU$iness Case: First Round many of the requirements needed for their ERP system, to include
e-business, inventoiy and MRP, and accounting.
Company production increased 500% in the first 2 years and steadily
At the price and .perceived functionality the decision to move
increased at about 20% each year thereafter. Early in 2003 it was
· forward with ERPlite was easy for the owners to accept. The risks
recognized that the old way of keeping track of sales, parts, and
seemed. very manageable, and if the 8ystem failed to deliver the loss
documentation was not adequate to keep up with the demand of its
would be mininiaL l1;1 · mid-2004, three .user seats were purchased,
customers. They began to explore sever~ information systems, includ-
with the intent to bttf more once the basics were in place and func-
ing but not limited to the followin~ tioning well. The three seats were installed, and everything worked
well for a couple of months. Testing went smoothly, but regardless
• In-house development
some limitations were discovered. The processing of work orders for
• High-cost vendor software ERP systems
the volume of work that WETI was experiencing not only took a
• Low-cost vendor systems
long time to compile but also created an overwhelming amount of
• Off-the-shelf software such as Q.uickBo()ks and Microsoft
paper trail.
Office Suite
Trade offs
In-house development was an option, considering the companies
experience in software engineering, but the limiting factor that pre- As the ERPlite systeJ:ll was altered to accommodate WETl's uni.que
vented development was time and manpower. Several midmarket ERP problems; it becani,e niore difficult for ERPli~e staff to provide effec-
software providers were· considered, such as Mi~rosoft Great Plains, tive over-the,-phone support, but not due to their lack of effort. Within
Sage software, and Infor ERP Visual. After several months of mov- a year WETI decided to forgo monthly support and decided it .would
ing through the typical sales routine, proposals were given to WETI, pay on a per-instance basis if further support were needed. Along with
50 SUPPLY CHAIN INFORMATION TECHNOLOGY
SUPPLY CHAIN MANAGEMENT SOFTWARE OPTIONS 51

··~d};~ceive ~~: 11 Alt-emati<tt¢s dimtdertd

·1a~~!l,t~i.Mim,t~ltatio9 .
Ovei: the pel'i~ 9f 3 months, ·keverafERP; ven~~ w~re ~u~:~
1~~:~1wl . . . ··~\~~:l~.o=~ti:;1 .'x.,,.,:. ·.
i;.. ;' ,. '
in,clu<.ling bunibtlimited to ~he,foUowir\$: ...
"··'·, •..>,,,,
· >
:.. {'./" '.'-;)if''
1
' '··
''.'·
'

~rk <>tder syst~

. . . ·. ,. :. !~.::;,,·;.;~'.· rqr . . ~'

• • Sage AccPac
t Epic<)/ .
• Mild4Manage •.
. Fo'1'~'la~r ttfC otf1"1l~~~ aBpe<:ts ERJ>l~'being·u~ ')Vere 4f •· Ef sh9p System
the pu~f\'ijsi~~,g.~t4f~na~~Q~.~es .m~µl¢,~tch~f1,g.~ld ~···· •. fucc JobBoss
.~pe stored ~~u¢t riumbers,;'iisip~f(tied supplier, &i4iidte most ~iii:
. . ... #ri~ t() generatetJ>ur~ase ordet'S. Hisr()ri~. f»rchase orde~< • Infor Global:s Vi~ual .·•.·. •.
• Mt by Bowen an~ Gr6ves (now Jig Ml)·
. sto~d i~th~~~'..~<j'.could be a~ed (or h~((>rkal ~ges·• .. ·d

i)1 J't,~~:1>,r i~Ji~~# $~~( ~u~d,p~dut~ ~<;iated with.~~p~riic

'the selecd9n Wa$lttarl'9wed to Visuhl attd Ml. TJiey; 1J:r>tll,~med
tar::·, pmject'or.pt9ductbuild.~;J~ would use' the.sxs~~m·to dc:term~"'.~
~ ' . :if.«. ; . ; .,,. .\\ {fil ··): : .' ; '. :>:':,,·;~'-,· '.'::',.~.\.~·':!)
to offer li comprehen$ive Sl)lution for~ engineeNJ,>"(>~tfi'.TQ) or
~ saleS pr(~ ~aSe4 on a .~talilatp markup of the put'Chase pri~. The
{ ',' ·di: 't ·'' ' ~ ·.;·~..,
" .. ··:·· ·.:.
1d'~ added t~lat~C$ <JllOfe at\(.l~nt ~l~ ~ O(~mai!.~ ·a'
. . . '~'.'''
::'· ·.~; . '. ', .. , :
m1de".~9"<>rder (~TQ) manufactµrin,12operation. Arttt an9thet c~~·
"'~... "f(l ~'*'~.th•t ~~ qutjf~ We~J{~~fy used ~t spafe pie .of rn,~:mths ·~fd~m~ristratlons with ,yarious :emplofa!$ of Wl!:Jl
/)} .. ,,'->-}£~; tt < . ~ >> > .~ ·;.
pims pnfy;~\not for Cr,ltire systems sales. ~µr<:h.ase orde~ . .
y:· . l,'' ' : ·. .
1m& more importantly With: the owriers::it .~decided ~bJ.if'fottrl
·
~ Sen~i'fecl ·~4 then printed to. pro~id~ 'accquriting ·.· posals fot a 1o~usef system with on·site frnplementation. ,uppot:i;1and
,3, ·,. ·'', '.' "':.':' '.'>.
adnn n~·td gene1'!lte a c;:~k from a~unts pay-:
. '.'. .·· ,. .. "f '
training for several:.end USC~f J"Jif bif!is raJ:lgec;i from $18iQQ(f~~ aQotlt
'»' *:---' :~'\.~~fir;:. .· ',.
. .,. . . .,.tee ~ttj~~~n~ tece!Vitble withi~VQ.uickBooks..Thf'S
·.: >•
. ; ",\ v.\ ··'. '···'' .- .,. ;.,." , . :'·'
$42,00Q with aef>roxima~Jy ~.n<a~i{ion.t! $5~()00 tO $~,09() CJpe~ .
icaeff~ of~ftort~ co~mo? in many art;as qfWETI's daily busi~ on a dediGl~ed ~~er and SQ.L datab~.40ftware. The owr,ters bal~
()~t1°:ps. '• : r . .. .. .. at• the price, anJlit was
quickiy teal~ ()n all sides that the oWn~,r~
had no intention ?f:.pµrchasing a S}'item a~ that paintin time;. The .·
Round two: Mfdmarket E~fl.y'endors Ve~sus Open-$durce company hac,i.just,h~lt a s~te·of-thc.-art ma11.1J.fac$ltriog ceil,ter,;alll
ERP i· . ·.:.< . wete heavily if!.vested in the development ofa llew a9to!ll.atj.Q,n {>ro~ .
uct watwas over hudtet and paSt the hoped~fur cofupletion ti~~ aqd,
The bu11!#~t~~ in thii ~,un~ 'proceeded qlu~h as in ~ht; ~tst round;
this oyers~dowed ~ny.,potc;llfml gai~ that woul!A· be realizecF(ro(ll a
This.fme•..~ever~~hosc(.in ,charge C>f finding the replaeemen~,for. ' ,''
purchaSed ERP ..System.
'· <, . ''

E~}f,~ stag~d with low-c:Ost ~lutions6rst~,1ooking at pr9grams such

·.~~:i~ ,Qa~~facturi~g .(i!~~:(/~.a~anuf,actti~i~~-~m) .. <tnd·· xTuple
. Fi~hbo~l)nv~fl>ry (h~~:l/~.6$hllowlinventQrf.C<>m). Th~.se Io~~~·:
~St soluti~~ W~re tesid ~~hgn free ~.tiaUs;~114,,w~ile 'ro9ltst ana Info;matj~n was ~inf en~l'.ed rito or, ,thr¢e tim~ in~S ~ l~~;
.wryfunctiPn'iu, t~ere ~rec6n~/ps voicedbv~t.;$~~'key lirili~atiOtJ 1 ' "systems" that each d~parrmeut ~ad ~ev¢1oped, mostly i~ &eel,;
..'~ng\VETI's;;~ode c;>f operlitions. . . . After about 3 ~ had ~'~mtfie d~~siQn·tO. ~.,~Atta;; ·
;~, . ,. ;''.f· , . ditional ERP, the head engineer approached those .involvedlwith th'C
ERP selection process and mentioned that his friend in a research and
SUPPLY CHAIN MANAGEMENT SOFTWARE OPTIONS 53
52 SUPPLY CHAIN INFORMATION TECHNOLOGY

dewto,..m;1a.,qf a... major ~ine~r w~. ~a:'prt>pm ~4

xTupl¢.t,c)"~~
·.··
of ptQR~t..: p~tiell. ~d costs. They
.. :.··.·.~··>•;- •..,,.,. . :• :. : :" . .. . . ; . . . ': ·. . . . : ..
were usRig g;t. ln,'¢njunqi,~n :Wi~ their 1;~~~ys ll!rfg¢,~endor
Ell> sya~m \o~-, ttack:of~ pro~rs inpltt$ and outputs, because 'J,

the la9.•,~~ was·~nly doing $0 from a macro view (slm-

ply ~·~lc·oftheoYe~depll,~ent$,.pll,tt$,.andcosts).·:w1ETl's
head. f!f~'11~-tlng ~· manli.f.lct~rinf~ ·~ed.,about t}ie product
d~l~::·c;:· ' " • .· ···. · · · k~j · fo . ··.·. ..to rJ.m,yitt1r CQlJ)P,liflf!: .·. .·
an . :. ~~<~'~'.W}t ~dle: a vt~ .~U,.tffJJ" r the: com- '"''V'···~.·<<
Y·' ··!~,, <:o·
.<'. .,.,
• , ·"';·;/'';· ·:.. '
t~ t~~jnS, it was ~1:4~ ~~·
pany's oiati:y)RP AeC<ft~ . . .
::-..:g;;;r~g~~·.
, , ,, , ' ' ~ " ' ,' ,"

~'l»P~ I~. ''"1

fil't~&¢d iti ·11: ~~~J~a11:•ed
die '$ttuctu~ro£,m~.ProgrQ#l and:the
After . ~. quick.;;~ of the,~fu~e·website,.'.the,, P4stbooks installer

!&t~~-
was,~~th. . ~~~fOip.~eiivia·.the xTuple ,Postbooks
pro~··H+:ln .,<ldit~~'•vdle soa~ ~. an SQL database was
also nce'd~lAH the ~Mt()!) info~tiiott,:including in£ormation
·~q·~'.SQ_Laa~'~tlll~tlon..•'.'.availa.le on:.d,le:·:x:Tuple /·~an.y'6f tht5~;:..,9f.pt0ducts,pnd·f~~;~ubcotnponen~ needed(gflrt'
~bsite. ~.4the o~rce database
··· <·.x · : .·.
:ii;J'uplein~s •al: .· n•lnim an<f~9s,which ad.Clea ~~~rabl}r w the·time it,.©okt<> .
. orga~~iAll':~!'~"4f~to• a ~r1jt;.at.;o,ui~ ~ .u$e{~r ~'Jhple~ \Viiliin·.·
with.
'ln. .tlon ~ - . ~a :ll ~-~:;of minutc:s the software was a'rew;.w~:'ill~ri~ ·!IP~mlfimatelf 15alifp l.OQQ0p;Jf~,'9~$ pf~yer
~·~~~onalin:.~···t·~if=~-~.nere was no <lataa~il~ · 1~,ooo)att~M.rnto tlteP~t1i!ooks ciatabase. · .·. • · · ·· .· ·. ·"' •
aDle,~ ~ti~ ws~m,_. tfi~m (4l aft~.ltn.~ tc:st~· There were~ 1s'f~c·<>1~<¥';fittfotril·~ ~owe4; for modificatif,)n of·thl ~~ctions •
few da~~ • dt~·hixi o~ dJe Sou~d>~ Postboriks. and retx>"~ ·t0~~¢iifi~:~· s?~re. ' f:his ~i'.~f~ies •t().
proje<:t p·/~n additiofl1-,~e empty~-~ there is also'd1e quick ixi<>ld'~·:~~\pund•soixi~~lflle busine$S'p~c:thilt:J>ro~de
start <lata-~ ~ .co~i•· a l>MiS:'chart of~nt$ and ~ tbe a ~mpe~ advantage: vyhik alscf providing the. benelirs of.a Jtill~ '·
account assiko~n\;S tequit«l t~·tl)n dull ~qc; of itansactions com- ·int.-ated·· E~ .progi;am. ~d.the best pr~tices •tnost ERP so~~ •is. 1
mon 10 ,~t' b~e.s. li;i adi"'«x
thete is a demo <lataba&e that
b~p~, ·'···· .' : . · ... · ·
containsi~ acci>untinl~ foonl'ib.:ihe qµi<'.k start <latabiise \md a
J .· .· .·· :: .••.•••·· .. ) : ·. :· • i : .•· · , · · • .. : ;

.: Ov.,.a~open,~1~-~~ llJeen pte\ien a cap.ble~Ju,:

set of •~ft,.<laca.slJdl. as p... accounting C>f accou~ti, bill of chart tion fut ~anr tiusi~~ 'i~ioµgh4~t th~"''*°rl,. ~ne· it may n~
mate(. ,Bf.JM) dat;t, and ~ on. For initial de11tonstf&tion Jnd test- ac~µ~.t tor·';'I• ~t•9f ritar~t .$h~rett~re··is demaridsfot .·
ing p~-~~'!lo~tabasr~!~· ·· .. such ~l~t~ni~·~a~~~~j~~ c;i~.51ip 1n~j ~Y;~~~a~ fro,nitlte ·.
,~;'initial .-me:.-tf:dte~ks edition of xTuple only attoptl~~;il~~,i~~~4)ij~~ftwa,er~~~wa>? f~.·liut·•·
lasted. a:~ ·~ t>y,the ~~n t. . . in .chatge:. ~finding and most coritp~nleit;,~~~~f;.~~~sfv~ts,ipn of ~~eir ~~·~with itri
impltm~"~~ for wt'Jl.~~~ 'N&S found tO bC very e&pll,ble a,nd upg~de.p:t)~:;~1~,re:~~;;~,i~·i~:n~~ . •.·. .·; c:~,.?
1• •.•.. ·
easy·'to·usef'i:tn:~·to··the·f.Vrsoiware. xTuple .also offe~1a
'•
...
&co' SQL rel'on.;~ ~~ of.4;;aris·feMD~·PfOYidc:s a very
'·!'···,· :: ,,·'« _,. , ..
Source: D. L Olson aM J. Staley (2012l:'
, • ''"''";, ~·· ·.· . §::,, ,
54 SUPPLY CHAIN INFORMATION TECHNOLOGY

Conclusion
There are many ways to obtain supply chain software support. Specialty CHAPTER4
products such as advanced planning systems are available. However, this
functionality is available in almost every ERP vendor system, with the
added benefit of integrating organizational computing. The amount of Business Process
money required to obtain software licenses is considerable, making it very
important for organizations to conduct a sound business case analysis. Reengineering in
This isn't easy, as there are so many options available. That is undoubtedly
why consultants are so often used (which by no means reduces expense!). Supply Chains
Two interesting options to avoid many of these pitfalls are outsourc-
ing to ASPs or using open-source software. Outsourcing has its own risks,
although a major benefit in up-front cash flow. However, outsourcing Michael Hammer coined the term business process reengineering {BPR)
hasn't been all that popular for smaller organizations. It appears more in 1990. Enterprise resource planning (ERP) systems depend on BPR
often for major organizations such as General Motors, Xerox, or U.S. to gain efficiency. The concept of BPR can be traced to its origins in
government agencies who wish to get out of the information system busi- management theories developed as early as the nineteenth century. The
ness and to focus on their key business operations. Open-source software purpose of reengineering is to "make all your processes the best-in-class."
has proven quite popular in Europe and South America, as well as for Frederick Taylor suggested in the 1880s that managers use process reen-
small organizations in the United States (especially state and local govern- gineering methods to discover the best processes for performing work
ments). These systems cannot be expected to provide all the functionality and that these processes be reengineered to optimize productivity. BPR
of well-tested commercial vendor products, but they offer sufficient func- echoes the classical belief that there is only one best way to conduct
tionality for many organizations. Installing an open-source ERP calls for tasks. Best practices are vendor methods selected to be the best way to
a new type of information system specialist, and this market is still under accomplish elemental business tasks. Accomplishing this goal depends on
development. maximizing the effectiveness of policy development and program deliv-
In chapter 4, we will look at the key task of business process reen- ery, planning and budgetary arrangements, decision-making processes,
gineering, followed by a closer look at business cases in chapter 5 and organizational structures, workplace relations, and people management.
software installation project management in chapter 6. Significant gains in performance have been attained through BPR.
BPR is suggested as a key step in the initial development of any orga-
nization's ERP. While best practices often are useful, organizations that
have developed core competencies in specific functions are better served
by avoiding the vendor best practices that all their competitors can buy
to retain what they do well. Amazon.com undoubtedly retains processes
that it developed to make it successful in e-business rather than adopt all
their ERP vendor's processes.
Increasingly, information and communications technology, often in
the form of an ERP, plays a vital role in determining the quality and acces-
sibility of services. Expansion in ERP has opened the door for even greater
56 SUPPLY CHAIN INFORMATION TECHNOLOGY BUSINESS PROCESS REENGINEERING IN SUPPLY CHAINS 57

efficiencies and enhanced service delivery through integrated processes. • Applying a spreadsheet algorithm such as exponential smooth-
The strategic application of ERP opens up opportunities for even greater ing over available monthly data
gains as public and private sector enterprises transform their existing • Incorporating seasonality indices into such a spreadsheet
processes. For individual enterprises and government agencies, the most algorithm
noteworthy gains are achieved when ERP decisions are business-driven. • Taking known orders and adjusting forecasts based on past
demand records
• Relying on managerial judgment
Processes
• Using guesswork (throwing darts, rolling dice, consulting a
A process is a logical set of related activities taking .inputs, adding value through Ouija board)
doing things, to create an output. In business, there are many different ways
to get work done. Information systems play a key role in providing a means to Whatever the forecasting method used, it can become part of the process
collect data, store it efficiently, generate reports to let management know what of determining the quantity to order each time an order is place.
the organization is doing, and archive data for future reference as needed. A business process is what the organization does to get its work done.
Blanket adoption of an ERP product will discard processes in which the For instance, supply chains need to take orders from customers and
organization has developed a competitive advantage. Instead of changing contact downstream sources to create the product to fulfill orders. This
those processes, the ERP system should be modified. Other activities will be process requires a number of actions. Before computer automation, this
better done following the ERP system's best practices. Even here, a transition could have involved a process given in Figure 4.1.
period can be expected where employees who have to radically change what
they do. Productivity degradation will occur while users learn to adapt to the
new system. In the long run the new system is most often better. Those who Customer places order

refuse to adapt to it usually have to learn new skills with their next employer.
In general processes can be divided into two broad categories. Opera-
Order sent to producer
tional processes have to do with accomplishing typical business functions,
including product development, order management, and customer sup-
port. Infrastructure processes are more administrative, such as establishing Producer c.lfders materials from sources
and implementing strategy and managing many aspects of the organiza- Source fills order, ships
tion to include human resources, physical assets, and information systems.
Producer recievcs materials
Each of these generic processes, whether operational or relating to infra-
structure, involves sets of tasks needed to accomplish work.
For example, in the operational process of order management, it is Producer makes product

necessary to forecast the volume of demand expected for the products

produced by an organization. The function of forecasting can be accom- Producer ships finished. product

plished in many ways:

• Using last month's demand as a prediction for this month Product received Producer bills Producer receives Producer sends
and pmccsst.-d cusromcr payment product tu customer
• Using the monthly demand from a year ago as the prediction
for this month Figure 4.1. Manual business process.
~·

58 SUPPLY CHAIN INFORMATION TECHNOLOGY BUSINESS PROCESS REENGINEERING IN SUPPLY CHAINS 59

The old process begins with receiving the customer order and then noti- The change implicit in BPR has many risks. Even advocates of BPR
fying the producer. The producer in turn then needs to notify its material cite failure rates of 50 to 70%. 1 Reasons for difficulty in implementing
sources to obtain the things needed to make the product. Often a single source BPR include the following:
was preferred, for simplicity and to reduce communication requirements.
Once materials are received, the producer can proceed with production and, • Employee resistance to change
when completed, ship finished goods to the core supply chain organi:zation. • Inadequate attention to employee concerns
This organi:zation then makes sure payment is received and, when this pay- • Mismatch of strategies used and goals
ment is confirmed, sends the product to the customer. The process is linear • Lack of oversight
and slow and has potential for miscommunication. Keep in mind that this is • Failure in leadership commitment
a simplified supply chain and could involve many other elements.
Utilizing information technology offers many opportunities to reen-
Best Practices
gineer processes, making them more efficient. A possibility is given in
Figure 4.2. One of the primary features of the SAP ERP product has been best prac-
Electronic automated systems can expand opportunities by reach- tices. Business process reengineering is an activity designed to identify a best
ing lower price sources from a larger pool of candidate sources. This also practice. Once a best practice is identified that would seem applicable to
can reduce risk because more alternate sources are available in times of most organizations, it can be incorporated into an ERP system. SAP spends
crisis, such as disruption by earthquakes, tsunamis, volcanoes, or wars. considerable research efforts to identify the best way of doing conventional
Deregulation arid competition are drivers for the creation of new business ERP tasks. They had 800 to 1,000 best practices included in their R/3 soft-
models through BPR. ERP systems provide a higher level of flexibility in ware. 2 Consultants often develop further specialized expertise that firms can
meeting growing customer demands, while demanding higher levels of purchase. A best practice is a method that has been judged to be superior to
automation and integration in almost all business processes. other methods. This implies the most efficient way to perform a task.
A related concept is benchmarking. Benchmarking is comparing an
Customer places order over Internet organization's methods with peer groups, with the purpose ofidentifying the
best practices that lead to superior performance. Best practices are usually
identified through the benchmarking phase of a business process reengineer-
Producer plant schedules ing activity. Best practices thus often change the organi:zational climate and
attempt to bring about dramatic improvements in performance.
Sources
Vendors attempt to be comprehensive and to be all things to all
send material people. But missed deadlines, excessive costs, and employee frustrations
Plant produces
are common in the implementation of ERP. A more participative design
Ii approach could help in implementing ERP. If a client implements the
entire suite of SAP modules, as well as their tools for system implementa- ·
Product directly tion, SAP can ensure timely implementation within budget. However,
this approach disregards the human factors of the client business culture.
While business process reengineeringwas designed to consider human
values and business purposes, these factors are clearly neglected in BPR
Figure 4.2. Automated business process. application. Care needs to be taken to consider human factors in new
60 SUPPLY CHAIN INFORMATION TECHNOLOGY BUSINESS PROCESS REENGINEERING IN SUPPLY CHAINS 61

processes. The human factor costs of training and obtaining cooperative does not have to be changed {it is the basis of the design). Cap Gemini
participation is the key to the successful implementation of ERP. refers to technology-enabled reengineering as concurrent transformation.
The technology-enabled approach designs the organizational sys-
tem around the abilities of the vendor software. SAP's best practices, for
Reengineering Options
instance, are designed to do things right in the first place. If SAP's research
Implementing reengineering can occur in two basic ways: either dean- came up with ways to do everything you do better than you used to do
slate or technology-enabled BPR. While these are not the only choices them, this would be the best option. It is the easiest to implement, is usu-
(they are the extremes of a spectrum of reengineering implementation ally much faster to implement, and thus costs less to implement. On the
possibilities), they are good concepts to explain the choices available in negative side, it also usually involves the most change in organizational
accomplishing reengineering. practice and thus the most complications for training. In practice, there-
fore, while the ERP installation project looks great from time, budget, and
functionality perspectives, the actual benefits to the organization are often
Clean-Slate Reengineering
disappointing. Table 4.1 compares trade-offs across these approaches.
In dean-slate reengineering, everything is designed from scratch. In essence,
dean-slate engineering involves reengineering, followed by selection of that Table 4.1. Comparison of Clean-Slate and Technology-Enabled
Reengineering
software best supporting the new system design. Processes are reengineered
Cle~·slate advantages Technology-enabled advantages
based on identified needs and requirements of the organization. As its
Not constrained by tool limitations Focus on ERP best practices
name implies, dean-slate reengineering has no predefined constraints. This
Not limited by completeness of best prac- Tools help structure and focus
theoretically enables design of the optimal system for the organization.
tice database reengineering
This approach is more expensive than technology-enabled reengineering,
Company may have unique features where Process bounded and thus easier
but dean-slate reengineering is more responsive to organizational needs. vendor best practices aren't appropriate
Clean-slate reengineering is slower and harder to apply than the Not subject to vendor software changes Know that design is feasible
technology-enabled approach to implementation. However, dean-slate reen- May be only way to embed processes like Experience of others ensures design will
gineering offers a way to retain competitive advantages that the organization web, bar coding into new technology work

has developed. Ideally, this approach can develop the optimal system for the Maintain competitive advantage Greater likelihood of cost, time
achievement
organization. Clean-slate reengineering can also involve significant changes
Software available (already developed)
in the way that the organization does business. However, the adjustment in
how organization members do their business often retains the features that
Clean-slate disadvantages Technology-enabled
were found to work well in the past. Thus, while training is required, the
disadvantages
impact is probably less than in the technology-enabled approach.
No preexisting structure to design Reengineering limited by tool
Greater likelihood of infeasibility System evolution possibly limited by
technology
Technology-Enabled Reengineering
May involve more consultants System evolution may be limited by
In technology-enabled reengineering, first the system is selected and then technology

reengineering is conducted (constrained reengineering). The reengineer- May be more costly, slower No relative advantage (others can pur-
chase same system)
ing process is thus constrained by the selected system. This approach is
May not work with selected ERP All best practices may not be available
faster and cheaper than dean-slate reengineering, because the software
62 SUPPLY CHAIN INFORMATION TECHNOLOGY BUSINESS PROCESS REENGINEERING IN SUPPLY CHAINS 63

Many organizations have difficulty in efforts to switch from old leg-

SAP representatives and dozens of Deloitte eonsultants partici-
acy systems to ERP. These legacy systems included distribution, financial,
pated in planning as well. The COOT project mana.ger met daily
and customer service systems developed in-house over the years.
with the COOT chief information officer (CIO) and updated the
project plan weekly. BPR addressed over 4,000 ·requirements for
BPR in the Colorado Department of Transportation
the system.
TJte Coku:ado. Oepart1Jlent of Transportation (COOT) is respon- An independent verification and validation firm was hired to
sible for building and maintaining Colorado's highway construction analyze Deloitte's processes to ensure adherence to the contract.
system. ·Their annual.•operating budget.is in the $1.5 billion range. This firm identified discrepancies that were reported to the CIO in
CQOT utilized computer inforIJ1ation systems for record keeping and December 2004. Concerns were expressed about defective require-
management~ However, a variety oflegacy systems had been developed ments due to insufficient requirements analysis and truncated l3PR
over the YC!.fi for spe:ific functions, and these systems were inconsis- efforts. There also was concern about testiu~ and quality assurance
tent and not. integrated. policies used by the illstallation contractor. It was suggested that
CDOT's Information Technology Office adopted a project to the ERP implementation be placed on hold until BPR issues were
replace theirJeg3fY information systems, which were considered to be resolved. However, the person making this suggestion was removed.
badly out of;~ate. They hited CIB£R, a consulting firm to '1SSe$S their from the implementation team, and the ERP implementation proj-
system. CIBER producecfan Information Technqlogy Sttategi<; Plan ect continued.
for COOT, concluding that COOT needed an off-the-shelfERP. at an The project consisted of phased introduction ofspecific ERl'
est~g'\ated ~ost of $50 million. COOT information technol<>S'f leaders modules. The Human Resources module was scheduled to be
sul>sequently adopted an e:x:rensive BPR study to evaluiite how candi- lauµched on April l, 2006, not including the payroll functi()n. This
date ERP pa¢kiiges would s<irveCDOT. launch succeeded. Rel'JlaiJ:}.ing modules, includii:ig payroll, were
By the end of 2003, the BPR project had sufficiently progressed to scheduled to launch November 1, 2()06. User acceptance tesu were
enable are<tuest for bids. J3ids werereceivedfromSiemens, :PeopleSoft, conducted before that Qare, and training was prqvide(i .to COOT
Oracle, andSAP. Siemens was eliminated from considetatipn, and employees. However, this training proved insufficient, and. the new
PeopleSo~ had been purchased by Oracle, leaving SAP and Oracle as system extensively increased the work the COOT employees ~dto
remaining bids . The two systems were carefully stored by a commit- do to enter payr:oll information. In addition to training insufficien-
tee pf a dozen COOT employees on criteria to include price, fit with cies, the new system was not extensively tested, and there was no
needs, a.nd ability to meet COOT functional requirements. The com- recovery plan.
mittee selected the SAP bid. The case demonstrates the role of BPR in extensive software imple-
Because instaU~ng an ERP. has proven complex, a consuhing firm mentations. Ideally, BPR should be conducted prior to software
experienced in SAP installation was sought. Bids in response to this selection and final design. Often, as appears to be the case at COOT,
request foi proposal were received from thtee consultants. Deloitte BPR is not adequately conducted. A survey of COOT staff in 2008
was selec~ed alter a rigor<;>us l'eview pr<;1Cess. indicated that 60% wanted to scrap (or fix) their I-year-old computer
·.The SAP installation project ini~iateQ in the summer of 2004, system.
led by· a stfering. committee of four full~time procurement. repre--
sentatives plus 60 representatives of COOT major functions. Two Sourer. Adapted from McCubbrey and Fukami (2009).
64 SUPPLY CHAIN INFORMATION TECHNOLOGY

Conclusion
BPR is an important philosophy. It aims to achieve improvements in CHAPTER5
performance by redesigning the processes through which an organiza-
tion operates, maximizing their value-added content. This approach can
be applied at an individual process level or to the whole organi:zation. System Selection
Business process reengineering is often a major component of an ERP
installation. This implies massive changes in the way in which organiza-
tions do their business. This has great potential payoff but also implies a Installing supply chain management (SCM) systems, whether a stand-alone
great deal of change in people's work lives, which requires a lot of atten- system or as a component of an enterprise resource planning (ERP) system,
tion to demonstrate benefits, as well as a great deal of retraining. would seem to be a simple matter. After all, the software, while expensive,
Requirements analysis is important in identifying what a proposed has been thoroughly tested by many customers. However, ERP implemen-
system is to do. In ERP projects, requirements analysis takes the form tations are consistently reported to have failure rates in excess of 60%. 1 A
of business process reengineering, to identify the best way (best prac- business case should be a key part of the process of selecting the form of
tice) for each business process supported by the system. Business process supply chain software for an organization. This chapter will demonstrate
reengineering can be accomplished in many ways, but two ways repre- fundamental means of financial analysis and show how other methods can
sent the extremes. Clean-slate BPR starts from scratch and is the ideal be used to consider other factors describing expected project performance.
approach. Technology-enabled BPR begins with the software selected. Cost-benefit analysis seeks to identify accurate measures of benefits
This is faster and less expensive, as many of the processes are selected and costs in monetary terms and uses the ratio of benefits to costs. Costs
from the system. In practice, neither extreme is necessarily best. Ham- and benefits for software systems are inherently difficult to estimate accu-
mer and Stanton credited reengineering as doing a great deal of good, rately because the many uncertainties, many hidden (unexpected) costs,
despite being a euphemism for mindless downsizing by some. 3 BPR has and potential benefits are difficult to measure precisely.
enabled companies to operate faster and more efficiently and to use infor-
mation technology more productively. Employees often obtain more
authority and a better understanding of the role their work plays for the Total Cost of Ownership
organi:zation as a whole. Customers get higher-quality products and more Different types of cost are unique to supply chain software applications:
responsive service. Shareholders obtain larger dividends and higher stock
values because BPR reduces cost and increases revenues. Executives no 1. Obvious cost-The license cost of the software itself is the most
longer see their organizations as separate entities but instead see them understandable.
as related elements in larger systems linked through information flows 2. Cost ofsystem integration--This is probably the single major factor in
across the business, reaching customers and suppliers. determining the long-term cost of a system. If a small company buys
an inventory control and accounting system, they would want the
system be interfaced with their customer relationship management
(CRM) system and also to their shipping system in the warehouse. The
company will also need same modifications to the central part order
entry system, which interacts with customer and shipping data. Inter-
faces and the customi:zation are costly and more involved with time.
SYSTEM SELECTION 67
66 SUPPLY CHAIN INFORMATION TECHNOLOGY

3. Cost of Implementation-This is the cost of getting the system live Cost-Benefit Example
in the first place. Implementation costs vary widely based on the
Consider a proposed supply chain software implementation involving an
application. Components of this cost are data migration services,
implementation study in Year 1 conducted by a small team of company per-
systems integration, training, consulting, process engineering, and
sonnel, aided by a hired consultant at $500,000 in the first year. This is a fairly
project management. small implementation, and at this stage of the analysis, typically expected costs
4. Expemes of Customization-Almost every business software imple-
are underestimated. This analysis in Year 1 includes some business process
mentation calls for at least some customization. As technology has
reengineering and the formation of a training team. This cost analysis assumes
advanced, the cost of customization has come down; however, this
the purchase of a $500,000 supply chain software system from a reputable
continues to be a major cost with supply chain software.
vendor, with maintenance costs for patches and upgrades of $100,000 each
5. Platform-Software requires a computer platform. The older the
year thereafter. Extra hardware to support the proposed system will be needed
computer platform, the higher the likelihood that a more powerful
in Year 1 at $750,000. Training expense will be low in Year 1, high in Year 2,
platform will be required to adequately support a new supply chain
and drop thereafter. The firm wants to treat software acquisition, hardware
software system.
acquisition, and consulting in the first year as an investment.
6. Safeguarding Costs-Maintenance to sustain the system is an ongo-
Operating the system will also incur_rosts. This budget is expected to
ing annual cost. Companies often charge around 20% of the
be $1 million in Year l, growing at a rate of 10% per year thereafter. The
purchase price per year for this cost category.
firm's cost of capital is 10%.
7. Training Costs-Training of user personnel is critical. A period
of about 1 year is usually required until the trauma of new system The board disagreed somewhat about the expected rate of growth of
implementation passes. This difficult period is easier to cope with if benefits. The dominant group on the board expects benefits from the
a good, thorough training program is adopted. Managers generally supply chain system to be $2 million per year in Year 2, $3 million in
tend to underestimate the magnitude required in such a training Year 3, and to grow at 30% per year thereafter. The time horizon selected
is 5 years, as new software is expected to outdate any currently available
program.
system after that time.
Total Cost of Ownership (TCO) has long been recognized as a sig- The board often disagreed about assumptions in cost-benefit models.
nificant factor in supply chain software strategies and decisions. Yet while A vocal minority chinks that this expectation is far too high and that ben-
both end users and vendors tend to talk about lower TCO, and many efits will only grow at the rate of 10% annual increase after Year 3. This
vendors claim it as a point of differentiation, seldom do they speak in data is summarized in Table 5.1.
terms of specific metrics. 2
Because supply chain systems involve long time frames (for benefits, Table 5.1. Cost-Benefit Analysis Input Data-Example Proposal
if not for costs as well), considering the net present value of benefits and Year Internal Consultants Software Hardware Training Benefits
costs is important. We will start with basic cost-benefit analysis and then team
1 1,000,000 500,000 500,000 750,000 500,000
consider value analysis (one way to deal with subjective benefits) and a
2 1,100,000 100,000 750,000 2,000,000
simple version of multiattribute utility theory (a means to consider trade-
3 1,210,000 100,000 1,000,000 3,000,000
offs across multiple measures of value).
4 1,331,000 100,000 800,000 3,900,000
5 1,464,100 100,000 400,000 5,070,000
Totals 6,105,100 500,000 900,000 750,000 3,450,000 13,970,000
SYSTEM SELECTION 69
68 SUPPLY CHAIN INFORMATION TECHNOLOGY

Cost-benefit analysis is somewhat arbitrary because it usually is Table 5.2. Net Present Value Calculation-Example Proposal
C/ • '.·

applied considering up-front expenditure as investment, used in the Year --:_ ~

Net cash flow Discoun~ ' .
1 -3,250,000 -2,954,545
denominator, with net cash flow thereafter used in the numerator. Con-
2 +450,000 +41,322
tinuing with our earlier example, investment here is the $2, 150,000
3 +690,000 +518,407
spent on consultants, software acquisition, and hardware. One version of
4 +l,669,000 +l,139,949
the cost-benefit ratio would thus be the following:
5 +3,105,900 +l,928,520
($13,970,000 = $6,105,100 - Total +2,264,900 +673,653
$3,450,000)/$2,150,000 = 2.053
For Year 1, the total cash flow (benefits minus all five expense catego-
This looks like a very profitable investment, with the investment cov-
ries) is -$3,250,000. That is equivalent in today's value to -$3,250,000/
ered 2.053 times. Furthermore, the arbitrary nature of what to include as
(1.1A1), or $2,954,545. For Year 2, the total cash flow is +$50,000,
investment can be demonstrated by assigning all costs to investment (not
equivalent in today's value to $50,000/(1.1 A2), or $41,322. In this case,
recommended, but for purposes of demonstration): over the 5-year planning horizon, nominal net cash flow is estimated to
$13,970,000/($6,105,100 + $500,000 + $900,000 + be a net gain of over $2 million. This stream of cash flow at the cost of
$750,000 + $3,450,000) = 1.193 capital (discount rate) of 1.1 is a net gain of $673,653, indicating that
at the cost of capital used, over the planning horiron considered, this is
Here benefits still cover investment, but the ratio is much lower.
a worthwhile investment. (The indicator is whether or not the sum of
Another problem is that the benefits are all in the future, while the bulk
discounted cash flows is positive or not.)
of the costs are early. Money has a time value. Note that if we increased the discount rate to 1.2, the NPV would be
a negative $221,234, meaning that the proposal would not financially
Net Present Value Calculation justify a cost of capital of 20% per year. This implies that the proposal
would break even if the project achieved somewhere between a 10% and
Net present value (NPV) is calculated by discounting the net cash flow a 20% rate of return. We can try various discount rates on a spreadsheet
in each period by the discount rate to the power of the time period and until we identify a discount rate just above 16.9% where the NPV will
then adding these discounted cash flows over the horiron of the analysis. be zero. The implication is that this investment would return 16.9% per
Table 5.1 organizes cash flows for each category by the time period used year on the investment.
in the analysis (in this case, by year). Note that a more accurate NPV can Cash flow analyses of this type are very useful, especially with spread-
be obtained using a shorter time period. Ultimately you could calculate sheet support. It dearly displays the many assumptions made. It is
NPV per second, but that is far too precise. Your savings account is prob- common for business decision makers to be uncertain about most of the
ably calculated on a daily basis. For purposes of analysis, a monthly period entries in the spreadsheet. In fact, it is common for different individuals
is probably good. But the shorter the time period, the more spreadsheet to have different beliefs about specific assumed values. The good thing
rows would be required, and the harder it would be to display. We will about spreadsheet analysis of expected cash flow is that each different set
use a year as our time horizon. of assumptions could be entc;:red, and expected NPV identified.
NPV is obtained by aggreg:uing each time period's cash flow and
~-_/
dividing by the discount rate to the power of the time period, as shown in
Table 5.2 using a discount rate of 1.1 per year.
70 SUPPLY CHAIN INFORMATION TECHNOLOGY
SYSTEM SELECTION 71

Payback Sensitivity Analysis

One of the most common reasons for company failure in the United A very good thing about spreadsheet cash flow models is chat you can
States is lack of cash flow. In our example, if the firm has cash flow dif- include any assumption. (On the other hand, the bad thing is that so
ficulties, the investment would be less attractive than if they had adequate many assumptions are required!) To demonstrate, we might assume
cash reserves. Making the assumption that over time cash flow will turn
a vocal minority on the board thinks chat the expected rate of benefit
positive at some point and remain positive thereafter, we can use another
increase is too high, and instead of 30% growth in benefits after Year 2,
metric, payback, to identify the time until the investment will be recov-
they expect that these benefits will only grow at the rate of 10% annually.
ered. This could be done in net present value terms (using discounted
We can demonstrate the impact of this assumption change in Table 5.4.
cash flows), or in nominal terms (with undiscounted cash flow). Eicher
(That's what sensitivity analysis is-checking for the impact of changed
way, payback is interested in how long it will take to recover investment.
entries into the model.)
The calculation is simply cumulative cash flow. Given the assumption of
Here the assumption in benefit growth in Years 4 and 5 doesn't change
an initial investment leading to future positive cash flows, the time until
cumulative cash flow turns positive is payback time. This is often used by the decision if payback is desired within 5 years. But the project would be

decision makers to gauge the attractiveness of an investment. Table 5.3 expected to be much less profitable (to the tune of $2 million in nominal
shows calculations for our example SCM system, for both nominal and cash flow). The NPV at 10% discount would drop to -$630,282, indi-
discounted cash flow. cating that with the lower benefit growth assumption, the project would
Here the payback period is under 5 years in both cases. Using nomi- not return 10% per year. The return on investment (ROI) turns out to
nal (undiscounted) values, it would take 4 years and just over 3 months be 1.021, or an average of 2.1 % return on the firm's money. They could
to recover the initial investment (assuming a linear rate of cash flow in probably find better things to do with their investment capital should
year 5). With the discounted numbers, it would take 4 years and almost 8 benefit growth be the more conservative.
months to recover the investment. In either case, the payback is between The problem with cash flow spreadsheet analysis is that practically
4 and 5 years. We can also see chat the firm will need cash reserves of at every entry is an assumption, which could be debated. As we said, the nice
least $3,250,000 to survive the project (nearly $3 million in net present thing is that it is easy to change and see the impact of any one assumption
value terms). (or set of assumptions). But there can be many details to check.

Table 5.3. Payback Calculation Table 5.4. Impact of Benefit Growth Rate
·· ..

Year Net Cumulative Net Cumulative Year Expenses High High Low Low
(undiscounted) (discounted benefits· . . . cumulative benefits cumulative
10%) 1 3,250,000 -3,250,000 -3,250,000
1 -3,250,000 -3,250,000 -2,954,545 -2,954,525 2 2,000,000 2,000,000 -3,200,000 2,000,000 -3,200,000
2 +50,000 -3,200,000 +41,322 -2,913,223 3 3,000,000 3,000,000 -2,510,000 3,000,000 -2,510,000
3 +690,000 -2,510,000 +518,407 -2,394,816 4 3,900,000 3,900,000 -841,000 3,300,000 -1,441,000
4 1,669,000 -841,000 +1,139,949 -1,254,866 5 5,070,000 +2,264,900 3,630,000 +224,900 +224,900
5 +2,264,900 + 1,928,520 +673,653 +673,653 Totals 11,705,100 13,970,000 11,930,000
72 SUPPLY CHAIN INFORMATION TECHNOLOGY SYSTEM SELECTION n

Value Analysis Table 5.6. Qualitative Features of ERP Options in Example

Alternative BPR Standardize Internet Advantage Keep Disruption
Peter Keen proposed value analysis as an alternative to cost-benefit analysis up
in the evaluation of proposed information system projects. These projects, A-Full Oimplete Complete Best Equal Best 5 years
clearly attractive to business firms, suffer in that their benefits are often B-FA,MM, Partial Partial Best Less than A Less 4 years
heavily intangible. For instance, decision support systems are meant to pro- SCM thanA

vide decision makers with more complete information for decision making. C-SCM Minimal Partial Best Less than B Less l year
than B
But what is the exact dollar value of improved decision making? We all
D-Inhouse Oimplete Complete Pmhlematic Best Mediocre 7 years
expect the success of firms to be closely tied to effective decision making,
E-Open Minimal Partial Good Moderate Low 2 years
but better decision making cannot be rationally and accurately measured. 3
source
Value analysis was presented as a way to separate the benefits measured Worst Worst 0
F-Nothing Nothing Nothing Worst
in intangible terms from costs, which are expected to be more accurately
measurable. Those tangible benefits as well as costs can be dealt with in Value analysis would consist of presenting the decision maker with
net present terms, which would provide a price tag for proposed projects. the intangible comparisons in performance and placing the decision in
The value of the benefits would be descriptive, with the intent of show- the context of whether or not the decision maker thought the improve-
ing the decision makers accurate descriptions of what they were getting,
ments provided by the new machine were worth their price tag. This
along with the net present price. The decision would then be converted
requires an analysis of expected benefits from each system. The fol-
to a shopping decision. Many of us buy automobiles, despite the fact chat
lowing are reasons this particular management team is interested in an
the net present cost of owning an automobile is negative. Automobiles
SCM:
provide many intangible benefits, such as making the driver look very
sporty, letting the driver speed over the countryside, and letting the driver
• To update current business processes (through business process
transport chose they would like to impress. The dollar value of these intan-
reengineering)
gible benefits is a matter of willingness to pay, which can be i~ntified ~
• To standardize procedures within the organization
monetary terms by observing the purchasing behavior of individ~is
• To make interaction with suppliers and customers over web
measurement requires some effort and is different for each individual.
technology possible
Assume a firm is considering five different ways to implement some or
• To gain strategic advantage
all of an SCM (plus the alternative of doing nothing). The options, with
• To keep up with competitors
estimated investments and benefits, are given in Table 5.5.
• To minimize system disruption
Table 5.5. Alternative System Implementation Options for Example ERP • To maximize positive net financial impact
Alternative Investment NPV
A-Full vendor ERP implementation, all modules $15 million $5 million The expected performance of each alternative on the six qualitative
B-Vendor ERP, only FA, MM and SCM modules $11 million $7 million factors is given in Table 5.6.
C-Vendor, only SCM module $8 million $8 million In chis approach, the expected benefits are understood to be highly
D-ln-house development $25 million -$2 million variable and are treated as a rough estimate. Costs are assumed to be a
E-Open-sourced system $3 million $4 million bit more reliable. Thus the options, in descending order of price, are as
F-Do nothing (current system) 0 0 follows:
.......

74 SUPPLY CHAIN INFORMATION TECHNOLOGY

SYSTE~ SELE~iJd I
A-Full implementation of vendor product expected to cost $15 million: £-Open-sourced system expected to cost $3 million:

• Complete BPR analysis and standardization • Minimal BPR analysis and standardization
• Top-of-the-line Internet access for suppliers and customers • Basic Internet access for suppliers and customer~
• State-of-the-art ERP system, which competitors also have • Proven ERP, but with less functionality than ot1'ers
access to • Minimal impact on operations
• Serious disruption of operations through installation in 5 years
Expected benefit: NPV $7, for net gain of $4 milli0
n.
Expected benefit: NPV $20 million, for net gain of $5 million.
F-Do nothing, expected to cost $0:
B-Partial implementation of vendor product expected to cost
• No BPR analysis and standardization
$11 million:
• Primitive Internet access for suppliers and custolners
• Partial BPR analysis and standardization • No ERP or SCM system, in market where competitors cl
• Top-of-the-line Internet access for suppliers and customers • No disruption of operations 0

• State-of-the-art for modules obtained, which competitors also

Expected benefit: NPV $0, for net gain of $0 millil)n,
have access to
• Serious disruption of operations through installation in 4 years
Management can now view each option as a market basket ~'I
Expected benefit: NPV $18 million, for net gain of $7 million. each with its own price tag. In this case, the key differ~11 • b()( 1:,jil
ce is ·jr
system in-house, adopting a variant of the vendor system, 1• all' ~1 itr
nst in r
C-Minimal implementation of vendor product expected to cost source software system, or doing nothing. Building £G.. g ,,jll
-. systell\ 11•
$8 million: dearly has many risks and involves the most out-of-pocket 1• .~
. . . nvestll\ t'
agement might well discard that option unless they ar~ ve e 1i
• Minimal BPR analysis, partial standardization
their ability to develop complex software projects. The oi-..e ry
~ n-sour f r·
COn~o~
• Top-of-the-line Internet access for suppliers and customers
• State-of-the-art for SCM only, which competitors also have
access to
has attractive features but involves more uncertainties ~d d ~ ~
the best BPR options. Among the vendor options, Option
features on reengineering, standardization, and Internet
:t\r
~I
~ ·~ l
• Serious disruption of operations through installation in 1 year connecrl'
it would involve 5 years of disruption, and call for a ~ 15 rnill' 1 ~~
Expected benefit: NPV $16 million, for net gain of $8 million. ment. Option B would save $4 million in investment wit~ the llJ~~tfl:
access at only 4 years of disruption but would sacrifice a ~it on~c ft\
D-ln-house implementation expected to cost $25 million: standardization, relative competitive advantage, and keeMn u ~~IA­
petitors. Management might feel that the gains in disrur..t. g p \\i .I\.
• Complete BPR analysis and standardization fO IOU are l ..
sacrifices in BPR, standardization, competitive advantag~ d "-' )r
wo4ida:v:tn~
• Suspect Internet access for suppliers and customers
ness. For only an $8 million investment, Option C
• Custom-designed system with features competitors don't have,
of even more BPR, strategic advantage, and competitiven.ess but c i~
but no best practices
3 additional years of disruption. However, it may be pa.._ ""0 h.
• Serious disruption of operations through installation in seven years "'amounttdr
higher degrees of methods improvement through busin~85 ro fr
Expected benefit: NPV $23 million, for net gain of -$2 million. neering and standardization of business functions acl'OSlt th~ o::/~'
76 SUPPLY CHAIN INFORMATION TECHNOLOGY
SYSTEM SELECTION 77

By focusing on the important features involved, management may be able to the simple multiattribute rating technique (SMART), an easy-to-use
conclude that Option A, B, or C is preferable to the other options available. method to aid selection decisions with multiple objectives.
Taking value analysis one more step, to quantify these intangible ben- Multiple criteria analysis considers benefits on a variety of scales
efits in terms of value (not in terms of dollars) takes us to multiple criteria without directly converting them to some common scale such as dollars.
analysis.
Multiple criteria analysis (with its many variants) is not at all perfect. But
it does provide a way to demonstrate to decision makers the relative posi-
Multiple Objective Analysis tive and negative features of alternatives and gives a way to quantify the
preferences of decision makers. 4
Profit has long been viewed as the determining objective of a business.
Fit with business procedures was selected among the three most
However, as society becomes more complex, and as the competitive
important criteria by about one-half of the respondents and was listed as
environment develops, businesses are finding that they need to consider
the single most important criterion by over one-third. While ERP ven-
multiple objectives. While short-run profit remains important, long-run
factors such as market maintenance, product quality, and development of dors have devoted a great deal of effort to making their packages match
productive capacity often conflict with measurable short-run profit. existing business processes, the importance of this criterion is based on
Conflicts are inherent in most interesting decisions. In business, the high cost and bother of configuring and implementing ERP systems.
profit is a valuable concentration point for many decision makers because Selection of a vendor involved less variance among criteria. Product
it has the apparent advantage of providing a measure of worth. Mini- functionality and quality were the criteria most often reported to be
mizing risk becomes a second dimension for decision making. Cash flow important.
needs become important in some circumstances. Businesses need devel- Perhaps the easiest application of multiple criteria analysis is the
oped markets to survive. The impact of advertising expenditure is often SMART analysis, which identifies the relative importance of criteria in
very difficult to forecast. Yet decision makers must consider advertising terms of weights and measures the relative performance of each alterna-
impact. Capital replenishment is another decision factor, which requires tive on each criterion in terms of scores. 5 We will first explain scores.
consideration of trade-offs. The greatest short-run profit will normally be Scores: Scores in SMART can be used to convert performances (sub-
obtained by delaying reinvestment in capital equipment. Many U.S. com- jective or objective) to a zero-to-one scale, where zero represents the worst
panies have been known to cut back capital investment in order to appear acceptable performance level in the mind of the decision maker and one
reasonably profitable to investors. Labor policies can also have impact on represents the ideal, or possibly the best performance desired. Note that
long-range profit. In the short run, profit will generally be improved by these ratings are subjective, a function of individual preference. Scores for
holding the line on wage rates and risking a high labor turnover. Some the criteria given in the value analysis example could be as in Table 5.7.
costs are not obvious, however, in such a policy. First, in a high-turnover
environment, training costs rise. The experience of the members of an Table 5. 7. Scores by Criteria for Each Option in Example
organization can be one of its most valuable assets. Second, it is difficult Option Bri. ~ •internet Advan'l·>
.... tat$ ·.~···.
. don ··~ f inanclal
for employees to maintain a positive attitude when their experience is that A 1.0 1.0 1.0 0.7 1.0 0.1 0.85
short-run profit is always placed ahead of employee welfare. And innova- B 0.9 0.7 1.0 0.5 0.8 0.3 0.90
tive ideas are probably best found from those people who are involved c 0.6 0.7 1.0 0.2 0.6 0.9 0.95
with the grassroots of an organization-the workforce. D 1.0 1.0 0.6 1.0 0.1 0.0 0.0
This variety of objectives presents decision makers with the need to E 0.6 0.7 0.8 0.6 0.4 0.8 0.7
balance conflicting objectives in ERP option selection. We will present F 0.0 o.o 0.0 0.0 0.0 1.0 0.2
78 SUPPLY CHAIN INFORMATION TECHNOLOGY
SYSTEM SELECTION 79

Table 5.8. Worst and Best Measures by Criteria

Table 5.10. Weight Estimation From Perspective of Least Important
Criteria Worst measure Best measure Criterion
Update systems (BPR) Nothing Complete Criteria Worst measure Best measure Assioned value
Standardize business Nothing Complete 7-Standardize business Nothing Complete 10
processes processes
Internet connectivity to None Modern 6-Financial Risk $25 mil, lose Risk $3 mil, gain 25
suppliers and customers implications $2mil $4mil
Gain strategic advantage Do nothing Develop unique system 5-Minimize disruption 7-year installation Current system 30
Keep up with cmnpetition Do nothing State-of-the-art vendor 4-Update systems Nothing Complete 50
(BPR)
Minimize disruption 7-year installation Current system
Financial implications Risk $25 mil, lose $2 mil Risk $3 mil, gain $4 mil 3-Internet None Modern 60
connectivity

Weights: The next phase of the analysis ties these ratings together into 2-Keep up with Do nothing Use
competition 70
an overall value function by obtaining the relative weight of each criterion. state-of-the-art
1-Gain strategic Do nothing
In order to give the decision maker a reference about what exactly is being Develop unique 100
advantage
system
compared, the relative range between best and worst on each scale for each
criterion should be explained. Many methods are available to determine these The total of the assigned values in Table 5.9 is 283. One estimate of
weights. In SMART, the process begins with rank-ordering the four criteria. A relative weights is obtained by dividing each assigned value by 283. Before
possible ranking for a specific decision maker might be as given in Table 5.8. we do that, we obtain a second estimate from the perspective of the least
To obtain relative criterion weights, the first step is to rank-order criteria important criterion, which is assigned a value of I 0 as in Table 5.10.
by importance. Two estimates of weights can be obtained. The first assigns the These assigned values in Table 5.10 add up to 345. The two weight
least important criterion 10 points and assesses the relative importance of each estimates are now as shown in Table 5.11.
of the other criteria on that basis. This process (including rank-ordering and
Table 5.11. Criterion Weight Development -
assigning relative values based on moving from worst measure to best measure Criteria Based on best Based on worst Comnromise
based on most important criterion} is demonstrated in Tables 5.9 and 5.10. I-Gain 100/283 0.35 100/345 0.29 0.33
strategic
Table 5.9. Weight Estimation From Perspective of Most Important advantage
Criterion 2-Keep 70/283 0.25 70/345 0.20
Criteria Worst measure Best measure Assigned up with 0.23
value competition
1-Gain strategic advantage Do nothing Develop unique 100 3-Internet 50/283 0.18 60/345 0.17 0.17
system connectivity
2-Keep up with competition Do nothing Use state-of-the-art 70 4-Update sys- 30/283 0.t 1 50/345 0.14 0.12
3-Internet connectivity None Modern 50 rems (BPR)

4-Update systems (BPR) Nothing Complete 30 5-Minimize 20/283 0.07 30/345 0.09 0.08
disruption
5-Minimize disruption 7-year installation Current system 20
6-Financial 10/283 0.04
6-Financial implications Risk $25 mil, lose Risk $3 mil, gain 10 25/345 0.07 0.05
implications
$2mil $4mil
7-Standardize 3/283 0.01
7-Standardize business Nothing Complete 3 10/345 0.03 0.02
husiness
processes processes

~
80 SUPPLY CHAIN INFORMATION TECHNOLOGY
SYSTEM SELECTION 81

The last criterion can be used to make sure that the sum of compro-
Supply Chain Software ERP Issues
mise weights adds up to 1.00.
Value Score: The next step of the SMART method is to obtain value Supply chain software can be very effective in managing large supply
scores for each alternative by multiplying each score on each criterion chains. Unfortunately, more than 90% of the companies that have imple-
for an alternative by that criterion's weight and adding these products by mented ERP systems have not had a truly successful implementation the
alternative. Table 5.12 shows this calculation. first time around. 6
These value scores (shown in the totals row) provide a relative score
that can be used to select (take the alternative with the highest value • Supply chain software should be motivated by accurate
score), or to rank-order (by value score). In this case, the SMART analysis strategic and tactical process improvement objectives, with
indicates a preference for Option A, the full version of the vendor ERP documented assumptions and valid ROI expectations and
system. This is followed relatively by Option B, which is to reduce func- metrics.
tionality to finance and accounting and materials management modules. • Supply chain software must be implemented appropriately and
Other options have lower ratings, while doing nothing is practically off in a timely manner to attain ROI expectations.
the chart in a negative way.
These two points may seem obvious, but supply chain management is
Table 5.12. Value Score Calculation not usually initially approached in this manner. As a result, many prob-
tijterta< W~pt. oPti6'1 Ol>tion Option Option Option· Opti~ lems come to pass during and after implementation. This often requires
,,,. ,<, ,,: ,.'
.· . A· 'a· c· D E F '< a reimplementation effort or at least a major tune-up. ROI comes from
Strategic 0.33 x 0.7 = x 0.5 = x0.2 = x 1.0 = x 0.6 = xO.Oc
process improvements that supply chain software supports, not from
advantage 0.231 0.165 0.066 0.330 0.198 0.000
new software itsel£ What's the difference? Software alone, no matter how
Compe- 0.23 x 1.0 = x0.8= x0.6= x 0.1 = x 0.4 = x0.0=
tition 0.230 0.184 0.138 0.023 0.092 0.000 good it is, has little impact on improving business performance. If you
Internet 0.17 x 1.0 = x 1.0 = x 1.0 = x 0.6 = x0.8 = x 0.0= continue to use presoftware business processes after implementation, you
0.170 0.170 0.170 0.102 0.136 0.000 can anticipate identical or possibly worse performance. Software can, on
Update 0.12 x 1.0 = x0.9 = x 0.6 = x 1.0 = x0.6= x 0.0= the other hand, enable many new processes.
(BPR) 0.120 0.108 0.072 0.120 0.072 0.000
Minimize 0.08 x 0.1 = x 0.3 = x 0.9 = x 0.0= x0.8 = x 1.0.
disrupt 0.008 0.024 0.072 0.000 0.064 0.080
ERP System Evaluation by Toroid lntematfonal Ltd.
Financial 0.05 x 0.85 = x0.9 = x 0.95 = x0.0= x0.7 = x 0.20.
0.043 0.045 0.048 0.000 O.D35 0.010
Adapted from H. S. <:. Pel"<lra and W. I<.,R. Costa (2008), The]QUrnal
ofBusiness Perspective 12(4).7
Stan- 0.02 x 1.0 = x0.7 = x0.7 = x 1.0 .. x0.7 = x 0.0•
dardize 0.020 0.014 0.014 0.020 0.014 0.000 Toroid International is a transfurmer m;inufacturer with roughly
Totals 1.00 0.822* 0.710 0.580 0.595 0.611 0.090 2,000 employees in Sri Lanka. They adopted an ERP system, based on
the decision of top- and mid-level management advised by their infot--
mation technology department. The original decision was based on
Toroid's business strategy, mainly driven by cost. A.de,tailed analysis of
impact on the company was not conducted. Options considered were
IFS, SAP, and Syteline ERP software packages. The origirud decisiofl
was supported primarily by vendor recommendations.
82 SUPPLY CHAIN INFORMATION TECHNOLOGY
SYSTEM SELECTION 83

Perera am« ~ptc>~.!l multiple ciiteria analrsis for ERP sys'.' The study confirmed the decision that Toroid had made, with the
tem $C~tW,,}. . . the: T.-~ cqe history ~ denionsttAte the u$C adopted Syteline system receiving the highest score. Table 5.13 demon-
of -~~~"~..{!\}iP) in tAis ~oncext. Judgniepts were . strates that Syteline's greatest advantage was in cost and benefits, which
based on i~~>~,,. . . ~p ~who had part~eipated.in had a relatively high weight.
the ERP i•p~~-~ Crlteria consi<l~ were as ful!Qws;
•Busi~~~~·. Conclusion
• hnt>~d S\J,pply dwn managemeQ:t · Many costs arise when implementing supply chain software. Many of
• I~~~~t~mersuPJ>ort these are hidden and especially difficult to estimate because most organi-
• Dec~risk zations don't repeat the exercise. We have reviewed some of the primary
• FuncW)n~l 6t
,, ':.,'"~""· .;{§S ','>,::,,
· methods used to evaluate SCM proposals. Cost-benefit analysis (with net
• T~hli~g'' present value used if the time dimension is present) is the ideal approach
• Ven<l,oi: ~Ition from the theoretical perspective but has a number oflimitations. It is very
• Cost atldbene6t5 difficult to measure benefits and also difficult to measure some aspects
• Ma~ap~~t ch~ges of costs accurately. One view of dealing with this problem is to measure
more accurately. Economists have developed ways to estimate the value of
AHP wa µse4.td generate weights for .the seven major criteria (the
a life and the value of scenic beauty. However, these measures are difficult
weight for l:tusi~.~~regy &elit into 0::75 for 8upply chain m~nager to sell to everybody.
nieht aapects Uti·.q.2;··fot in~teasingcustomer support). AffP was
Cost-benefit analysis provides an ideal way to proceed if there are
also used tac ~re -.;h of~e ~ candidate padciges on these seven
no intangible factors (or at least no important intangible factors). How-
criteria. ~ua,tiion · was ~ndiicted using Expert ·Choice s0fcware.
ever, usually such factors are present. Intermediate approaches, such as
Table 5;138f~~~ldna~r¢'s and\\tei.f\u.
'' , ,' ' ' payback analysis and value analysis, exist to deal with some cases. More
Table S.ll• Toroid.Int~~ J:;RP~cPcm Dota complex cases are better supported by multiple criteria analysis. One of
the most obvious difficulties is that benefits, and even costs, can involve
high levels of uncertainty. The element of chance can be included in cost-
benefit calculations by using Monte Carlo simulation. 8
Value analysis is one such alternative method. Value analysis isolates
Change I -0.0iO I 0,311 I 0:266 •I O.•m intangible benefits from those benefits and costs that are more accurately
management
measurable in monetary terms and relies on decision-maker judgment to
Risk .. 0.040 ·0.259 0.416 o.334
come to a more informed decision. The SMART method, one of a family
Fu~t.(61: 0.149 . 0.360 0.345 . 0.295
.. .. of multiple criteria decision analysis techniques, provides a way to quan-
Cost~· o.i8s ·< . '.'-0.~71 ins: 0.594
tify these intangible factors to allow decision makers to trade off values.
-
benefit&
o.333• Or333 0.333
.. o.-409' 0,305· 0.285

.o:~:zq' 0.287 I 0.392

CHAPTER6

Supply Chain Software

Installation Project
Management
Supply chain management support can be obtained many ways, some
of which were discussed in chapter 5. Regardless of how a supply chain
management (SCM) system is acquired, installing it is a major project.
Project plans need to be aggressive, because SCM systems are very critical
to organizational efficiency (the sooner the system is brought online, the
sooner benefits are obtained). It is difficult to reliably estimate durations
of any information systems project. Due to the need to balance realism
with urgency, schedules should be aggressive but achievable.
The type of SCM system adopted will impact the need for changes.
Adoption of a single vendor package without modifications will reduce
the need for customization, and thus will reduce project complexity,
which fosters better schedule performance. However, the primary objec-
tive is to develop the right system for the organization. Implementing
the enterprise resource planning (ERP) system on time and on budget is
important, but it is more important to implement the right ERP system.
A project management steering committee is very important to
ERP implementation project success. This committee should include
senior management representing affected corporate functions, as well
as representative end users. This committee usually is involved in sys-
tem selection, as well as monitoring project progress and management
of external consultants. Consultants provide experience missing from
the organization and can be invaluable. However, consultants should be
screened to avoid those with financial ties to sofrware vendors and lack of
experience in the specific system being implemented.
86 SUPPLY CHAIN INFORMATION TECHNOLOGY
SUPPLY CHAIN SOFTWARE INSTALLATION PROJECT MANAGEMENT 87

Perhaps the single most decisive element of SCM installation success

the rest of the organization. The thought is to prioritize the
or failure is the knowledge, skills, abilities, and experience of the project
functional areas and to bring into operation that area offering
manager. An SCM installation project manager must understand both
the greatest advantage first. This requires an enormous amount
the business and the technology. Private and public sector organizations
of interface programming to preserve the data flows between
are increasing their use of enterprise information systems (EISs), which
the legacy system and the new module being implemented. It
make integrated financial and human resources solutions available that
is also the lowest risk alternative but also takes the most time as
facilitate public sector business processes. While procuring the right EIS each module is rolled out.
system is an important factor, the genuine challenge lies in implementing
• Big Bang implementation approach--This requires simultane-
these systems, which can be tremendously complex. Today's EIS project
ous implementation of multiple modules of an EIS package
managers are faced with overwhelming challenges. As government and at one time. Why is it called Big Bang? An organization
private system users' budgets have become leaner, project managers have
prepares, tests, trains, and everything else needed to get ready,
to achieve more with less in tighter time frames. This chapter addresses
and then over a weekend or a few days the data in the old
the explicit needs of the private/public-sector EIS project manager, with system is migrated to the new one. One morning everyone in
wide-ranging know-how in private/public-sector project management. the company starts using the new system and the old one is
Project management is one of the most important fields in informa-
simultaneously turned off. This is the most risky alternative.
tion systems. It is difficult to bring an information systems project to
There will always be unforeseen and unexpected events. Several
completion on time, within budget, and meeting specifications. Reports
famous companies have been caught in this trap. Usually high
issued by the Standish Group based on surveys of IT companies repeat- technology companies that thought it could not happen to
edly find that only about 15% come in on time and within budget. For
them found that it could. A variant of the Big Bang approach
large companies, the success rate is lower. While EIS systems involve a
is to combine it with a phased approach. This entails a series of
little more structure than the average IT project, there still are problems
"minibangs" that affect a logical segment of the business. One
encountered in implementing EIS. A typical problem is underestimation example uses a division-by-division approach where each one
of the time to get an EIS system working. However, there have been a uses a Big Bang to migrate to the new ERP system. A second
number of efforts to make this type of installation less problematic. One example might use a functional approach, on the other hand
major reason is that vendors have a vested interest in making EIS installa- this requires interfaces with both systems running their parts
tion less risky and more predictable. of the company, that is, finance goes first with the new system
implemented in all divisions at one time, followed by manufac-
Implementation Process turing and customer support.
• Phased implementation approach--consists of designing, devel-
Three broad approaches to the timing and location of EIS imple-
oping, testing, and installing different modules of the EIS over
mentation include the pilot implementation approach, the Big Bang time or rolled over different elements of the organization at
implementation approach, and the phased implementation ilpproach: different times.

• Pilot implementation approach--This approach starts with a

small-scale version applied to a small division or with one spe- Implementation Issues
cific module, such as finance and accounting, with the intent
ERP solutions ultimately help in ensuring that data is transparently avail-
of seeing how that initial effort works before jeopardizing
able across the enterprise. The advantages of implementing a good ERP
88 SUPPLY CHAIN INFORMATION TECHNOLOGY
SUPPLY CHAIN SOFTWARE INSTALLATION PROJECT MANAGEMENT 89

system are manifold. With its integrated applications, an ERP system

ERP/SCM Installation Project Risk
optimizes the core processes of an organization, accelerates transactions
with its business workffow, and makes strategic management informa- Managing risk on an EIS project is crucial to its success. A risk is a
tion available in a transparent form at all levels of the company, within potential failure point. An EIS project may have thousands, maybe even
the framework of its information warehouse concept. An ERP system millions, of potential failure points, in the form of untested technology
guarantees strategic freedom in designing an organization's information (and untested staff}, political events, and even nature. When computers
interact with how humans do their jobs, often unintended consequences
management communication. This enables uniformity in storage and
occur. We can broadly categorize steps to manage risk:
restricted exchange of business data between two physically divided appli-
cation systems. Ready-to-wear solutions for more than a few hundred 1. Find potential failure points or risks.
business processes are available in an ERP solution that can be promptly 2. Analyze the potential failure points to determine the damage they
constructed using its business process such as modeling tools to optimize might do.
the processes of an organization. With its business-oriented functions, an 3. Assess the probability of the failure occurring.
ERP solution opens up new opportunities for improving an organization's 4. Based on the first three factors, prioritize the risks.
market and customer orientation through data mining techniques. The 5. Mitigate the risks through whatever action is necessary.
high level of flexibility is easy to get through ERP systems and enables the
flexibility to respond to a continually changing market situation. Project team members must rely on their experience and advice from
Where vendor software installation money is spent is of interest. In others to find potential failure points or risks. Track through the entire
project plan and look for areas of ambiguity:
the ERP field, two surveys 1 in manufacturing industries found the results
shown in Table 6.1.
Step I: One of the easiest and most effective ways to find poten-
In the United States, vendors seem to take the biggest chunk of the
tial failure points is to talk to other organizations that have
average implementation. Consultants also take a big portion. These
done the same projects. Cost estimates are probably the most
proportions are reversed in Sweden. The internal implementation team
common potential project failure point. Other potential failure
accounts for an additional 14% (12% in Sweden}. These proportions are
points include lack of an executive sponsor, an under qualified
roughly reversed in Sweden with training.
project manager, and no clear objectives for the project.
Step 2: The next step is to determine the severity of the poten-
Table 6.1. EIS Installation PToject Cost PToportions
tial failure on the budget, project timeline, or the users'
lnsta~.. ~st ·. \J~St:Jtes Sweden requirements.
propord09 .; ....
Step 3: Assessing the likely impact and the probability of the
Software 30% 24%
failure occurring is more art than science, requiring in-depth
Consulting 24% 30%
knowledge of both the EIS package and the business. A risk
Hardware 18% 19%
management team should be built that brings together those
Implementation team 14% 12%
individuals that have the knowledge and experience to know
Training 11% 14%
what might happen. This team must have experience in
Other 3% 1%
implementing the specific EIS package for an organization
Source. Extracted from Maben et al. (2000); Olhager and Selldin (2003).
approximately the same size and in the same industry as yours.
90 SUPPLY CHAIN INFORMATION TECHNOLOGY SUPPLY CHAIN SOFTWARE INSTALLATION PROJECT MANAGEMENT 91

Step 4: Based on the first two factors, prioritize risks. Decide Table 6.2. Activity List for Full EIS System
which risks should be eliminated completely, because of poten- Activity ~ration Result
tial for heavy impact on critical business processes. Sec up a A-Select consultant for selecting system 2 weeks none

monitoring plan for risks that should have regular management B-Meet with consultant to discuss ERP 4weeks A

attention. Make the entire team aware of those risks. C-Meet with Board of Directors to explain l week A
proposal
Step 5: You mitigate risks by reducing either the probability or the
D-Develop request for proposals to give to 3 weeks B,C
impact. The probability can be reduced by action up front to vendors
ensure that a particular risk is reduced. The project risk plan E-Ohtain vendor proposals 4weeks D
should include a sec of steps to recover from each risk, should F-Develop organization business cases including 8weeks E
failure occur. The team must know the person accountable for training

recovery from each specific risk, and the action to be taken to G-Present business cases to Board of Directors l week F

resolve it. The team must know the symptoms of the impend- H-Redo estimates in light of hoard requests for 3 weeks G
changes
ing failure, and act to prevent it from occurring if possible. An
I-Initiate installation project l week H
example is to test a particular operating system or hardware
component to prove that it works prior to going live. Doing a
pilot implementation or prototyping the first set of EIS inter-
faces are both examples of risk mitigation. 8-+8
Project Management Figure 6.1. Network for ERP system selection scenario.

To demonstrate the relative aspects of installing various forms of ERP/ assumes. For instance, analyses could anticipate and begin work on activ-
SCM, we will start with a simple critical path model (CPM) representing ity D before the board approved the concept. Such an early start would
selection of the SCM variant for an organization. Microsoft Project uses run the risk of wasting time if the board were to disapprove the proposal.
the traditional CPM, which consists of a list of project activities, each of But if management considered this to be of low likelihood, and if the
these activity's duration, and immediate predecessors. Microsoft Project benefit of getting a jump start on estimation seemed worth the risk, mod-
allows extending project management to resources required. Inputs are ifications to the implied set of precedence relationships could be made.
given in Table 6.2. It is necessary to understand specifically what the critical path method
This is a fairly straightforward sequence of activities, which can be assumes in order to make use of its output.
graphically displayed in a network, as shown in Figure 6.1. Networks also Another graphical outcome of the critical path method is a Gantt
provide a valuable visual aid for managers to identify relationships among chart, which displays the early start schedule versus time (Figure 6.2). A
activities. schedule can be generated over a period of weeks based on the early start
The sequence of events here is clear-activities B and C can work implied for all activities.
in parallel once activity A is accomplished. All other activities are Figure 6.2 gives the Microsoft Project Gantt chart for this project. A
sequential-they can't begin until their predecessors are finished. A Gantt chart shows the time periods when each activity is scheduled to
predecessor relationship is captured by the critical path method. In real- occur. The critical paths (project activities that muse be completed on
ity, these relationships are often not as rigid as the critical path method time if the overall project is to finish on time) are identified by a color.
92 SUPPLY CHAIN INFORMATION TECHNOLOGY SUPPLY CHAIN SOFTWARE INSTALLATION PROJECT MANAGEMENT 93

Late Start Schedule-. The next phase of the critical path analysis is the
late start schedule. The late start schedule is the latest an activity can be
scheduled without delaying project completion time. The final ending
time for the project can be some contract deadline, which may be differ-
ent from the early finish schedule, or the early finish project completion
time can be used. If the deadline is earlier than the project early finish
time, the project is infeasible (can't be completed on time with given
durations). If the deadline is later than the project early finish time, all
Figure 6.2. Schedule for ERP system selection scenario. activities in the project will have slack or spare time. If the deadline coin-
If everything went according to plan, this project would be completed cides with the project early finish time, there will be at least one critical
in 26 weeks. The activity list, estimated durations, and predecessor rela- path, connecting activities in a chain with zero slack.
tionships are all the information items required to develop a critical path The late start schedule is calculated in reverse. Networks are not really
model. That also means that all other information like uncertain dura- needed for development of early start schedules but are very useful in sort-
tions and limited resources are assumed away in the critical path method. ing out the relationships for late start schedules. Begin with the end time
The initial step of the critical path method is to generate the early start (deadline or early finish time). All activities that do not appear on the list
schedule. of predecessors for unscheduled activities can be scheduled. The late finish
Early Start Schedule: For every activity that has no unscheduled prede- time will be either the project end time or the minimum of the late start
cessors, schedule the activity to start as soon as possible (either the project times for all following activities. While networks usually aren't required to
start time--usually time 0 for reference-or the maximum early finish of do an early start schedule, they can be handy in sorting out the abstrac-
all predecessors). The critical path schedules are optimal with respect to tion involved in working backwards during development of the late start
time. This process continues until all activities are scheduled. The early schedule. The late start schedule for our scenario is shown in Table 6.4.
finish is the sum of the early start time plus the duration as shown in
Table 6.4. Late Start Schedule for ERP System Selection Scenario
Table6.3:
Activity Duration fQllowel'$ Early Early hsult
.. finish start .
Table 6.3. Early Start Schedule for ERP System Selection Scenario
I 1 week None 26 25 Releases H
Activity Durad& Predec~limts ·,;.,-'.'Early
a~th'
Result
>
,.';·.,,.<,
H
.l:J·tlJh
3 weeks I 25 22 ReleasesG
.. r :~: ...·. :'..Start ~ ,< ,,. , ,

G 1 week H 22 21 Releases F
A 2 weeks none 0 2 Releases B, C
F 8weeks G 21 13 Releases E
B 4 weeks A 2 6 ReleasesD
E 4weeks D 13 9 Releases D
c 1 week A 2 3 Must be done to
start D D 3 weeks E 9 6 Releases B, C
D 3 weeks B,C 6 9 Releases E c 1 week D 6 2 Releases A
E 4 weeks D 9 13 Releases F B 4weeks D 6 2 Must be
F 8weeks E 13 21 ReleasesG started before
A can finish
G 1 week F 21 22 Releases H
A 2 weeks B,C 2 0 Late start
H 3 weeks G 22 25 Releases I
schedule
I 1 week H 25 26 Finishes project done
94 SUPPLY CHAIN INFORMATION TECHNOLOGY SUPPLY CHAIN SOFTWARE INSTALLATION PROJECT MANAGEMENT 95

Here the only case where there were multiple early starts to compare Table 6.6. Activity Ust for Supply Chain SoftwaTe Installation
was for activity A. Activity /'ls early finish was the minimum of the early .A~lNitv'
.. ,, ' .., ·~tl'¥'· ··;:,,1. \;:J~ 'I) ;)' '>
'

t' ,,.,d:uta'tion.';' 'Jt~~~$Qrs

.,
,,,,.,.'
start for all activities that were followers of activity A (in this case activi- 1. Meet with consultant none
ties B and C). The minimum early start for those two activities was 2, so 2. Obtain vendor (provider) proposals 1
the earliest A could finish without delaying the overall project would be 3. Business case 2
the end of Week 2. 4. Obtain board approval 0 3
Slack is the difference between the late start and early start schedules 5. Form internal team 2 weeks 4
(it doesn't matter which you use, because in both cases the difference 6.BPR 4
between start and finish is the duration). Those activities with zero slack 7. Develop sandbox 4
are critical. If they are delayed, the project completion time would be 8. Train superusers 7
delayed. There can be more than one critical path for a project, and the 9. Set vendor software parameters 6
project network presented in Figure 6.1 can be useful in ensuring identi- 10. Phase 1 Installation 5,9
fication of each critical path. Table 6.5 shows identification of slack. 11. Phase 1 Testing 10
Slack in the case exists for only one activity, C. One critical path of 12. Phase 1 Load data 11
activities, consisting of the chain of activities from A to I, has zero slack. 13. Phase 1 Go live 0 12
More complex projects will include slack for multiple activities. 14. Phase 2 Installation 13
15. Phase 2 Testing 14
16. Phase 2 Load data 15
CPM Models of Supply Chain Software Options 17. Phase 2 Go live 0 16
In chapter 5, we considered a variety of means to obtain supply chain 18. Phase 3 Installation 17
software. The first option was to install a complete EIS vendor system, 19. Phase 3 Testing 18
including an SCM module. Table 6.6 gives a general list of activities, 20. Phase 3 Load data 19
some of which might not be used by particular options, but which serve 21. Phase 3 Go live 0 20
22. Phase 1 Train users 8, 11
as a general installation project framework. Durations would depend on
23. Phase 2 Train users 8,15
Table 6.5. Calculation of Slack for ERP PToject Selection Scenario 24. Phase 3 Train users 8, 19
Activitv,.~'.~v':' ·.~.,.tv . U..a·'···· Late · .t· Sl3ck I Critical?
Jsei· .inl$h I.l'tart ~b'
the option selected. Milestone activities all have zero durations. Form-
A 0 2 I o 2 0 Yes ing the internal team would take 2 weeks regardless of which option was
B 2 6 I 2 6 0 Yes selected. All other durations vary by the specific option.
c 2 3 I 5 6 3 No Duration data in Table 6. 7 roughly demonstrate time differences in
D 6 9 I 6 9 0 Yes the options discussed. Developing an in-house or open-source system
E 9 13 I 9 13 0 Yes involves more internal development.
F 13 21 I 13 21 0 Yes Activity 4 is an example of a milestone, an activity with zero duration.
G 21 22 I 21 22 0 Yes
This is an event, in this case triggering the development of an employee
H 22 25 I 22 25 0 Yes
training program for an organization adopting the system. The network
25 26 I 25 26 0 Yes
for this project is given in Figure 6.3.
96 SUPPLY CHAIN INFORMATION TECHNOLOGY SUPPLY CHAIN SOFTWARE INSTALLATION PROJECT MANAGEMENT 97

Table 6. 7. Acti'1ity Duration.• b:Y Option The Gantt chart from Microsoft Project for the full vendor system
Activity Full 3·module SCM In- Open ASP with three phases is given in Figure 6.4, with the critical path displayed in
ERP vendor only house source
red in Microsoft Project (dark gray in Figure 6.4).
vendor 1y1tem
system Project slack is displayed with bars for noncritical activities. These
1. Meet with 1 week 1 week 1 week slacks can be calculated in Excel as shown in Table 6.8.
consultant
In this case, one critical path consists of activities 1 to 6, 9 to 13, and
2. Ohtain proposals 4 weeks 4 weeks 4 weeks 1 week
3. Business case 10 weeks 8 weeks 5 weeks 10 weeks 5 weeks 5 weeks Table 6.8. Calculation of Slack for SCM Only System
4. Ohtain hoard 0 0 0 0 0 0 Activity Duration Early Early Late Late Slack Critical?
approval .·
start finish start finish
5. Form internal 2 weeks 2 weeks 2 weeks 2 weeks 2 weeks 2 weeks
I. Meet I 0 1 0 I
team 0 Yes
with
6. BPR 12 weeks 9weeks 5 weeks 15 weeks 8 weeks
consultant
7. Develop 5 weeks 4 weeks 2 weeks 3 weeks 2 weeks 5 weeks 2. Ohtain 4 I 5 I 5 0 Yes
sandhox
proposals
8. Train superusers 6 weeks 5 weeks 4 weeks 1 week 3 weeks 6 weeks 3. Busi- 5 5 10 5 10 0 Yes
9. Set vendor 15 weeks 12 weeks 6 weeks 1 week ness case
parameters
4. Ohtain 0 10 10 10 10 0 Yes
25. Development 26weeks 5 weeks hoard
10. Phase 1 3 weeks 2 weeks 1 weeks 6week 2 weeks 1 week approval
Installation 5. Form 2 10 12 19 21 9 No
11. Phase 1 Testing 4 weeks 3 weeks 2 weeks 12 weeks 5 weeks l week internal
12. Phase l Load 3 weeks 2 weeks 2 weeks 2 weeks 3 weeks 3 weeks team
data 6.BPR 5 10 15 10 15 0 Yes
13. Phase l Go live 0 0 0 0 0 0 7. Develop 2 10 12 20 22 10 No
14. Phase 2 3 weeks 2 weeks sandhox
Installation 8. Train 4 12 16 22 26 10 No
15. Phase 2 Testing 4 weeks 3 weeks superusers

16. Phase 2 Load 3 weeks 2 weeks 9. Set 6 15 21 15 21 0 Yes

data vendor

17. Phase 2 Go live 0 0 parameters

10. 1 21 22 21 22
18. Phase 3 3 weeks 0 Yes
Installation Phase 1
installation
19. Phase 3 Testing 4 weeks
11. Phase 2 22 24 22 24
20. Phase 3 Load 3 weeks 0 Yes
I testing
data
12. Phase 2 24 26 24 26
21. Phase 3 Go live 0 0 Yes
I load
22. Phase l Train 3 weeks 3 weeks 3 weeks l week 3 weeks 3 weeks data
users
13. Phase 0 26 26 26 26
23. Phase 2 Train 3 weeks 3 weeks 0 Yes
1 go live
users
22. Phase 3 26 29 26 29
24. Phase 3 Train 3 weeks 0 Yes
I train
users
users
SUPPLY CHAIN SOFTWARE INSTALLATION PROJECT MANAGEMENT 99
98 SUPPLY CHAIN INFORMATION TECHNOLOGY

We ran Microsoft Project models for each of the six systems. Com-
parative results are shown in Table 6.9.
Table 6.9 demonstrates a bit of the trade-off among different sys-
tems. More complex systems obviously are going to require longer
installation projects. ASP projects are probably going to be the fastest
to install. In-house systems can usually be expected to take the longest,
Figure 6.3. Network for framework with one phase.
although here our numbers show the full vendor ERP to take slightly
22. It is possible to have parallel sets of critical activities in more than one
longer, due to predecessor relationships imposed on the phases. There
path. Looking at the list of critical activities will not automatically identify
will be far greater duration uncertainty in the in-house and open-source
the critical path. This list needs to be considered in light of the network.
options.
Shared slack and independent slack are the two kinds of slack. Shared slack
is slack that is shared across more than one activity. For instance, both Projects have a bias in that activity delays accumulate, but any activ-
Activities 7 and 8 have 10 weeks of slack. This implies that some spare ities that finish early have to wait on other (slower) activities. 4 This is
time is available in the accomplishment of these activities. However, if a because when an activity is late, those that must wait for it to be com-
week of delay were encountered in Activity 7, this would reduce the time pleted start later than scheduled. On the other hand, if an activity should
available for Activity 8 as well. That is because it is the same 10 weeks of be finished before it was scheduled to finish, the advantage rarely can be
spare time in this case. Activity 5 has 9 weeks of independent slack. used, because in complicated projects different crews and materials have
Schonberger recommended focusing on scheduling the critical chain to be gathered, and many different people need to be coordinated. The
of activities closely, to make sure that they have the resources needed to
early finish time is not usually known much prior to the activity's com-
proceed as scheduled. 2 Goldratt adopted the same view. 3 The critical chain
pletion. Therefore, it is very difficult to gather all the following activities'
of activities includes those activities that are critical (as long as managerial
resources together in time to start early.
control can influence their duration) but is not limited to these activities.
Activities with very little slack can become problems if they are delayed up The following case demonstrates some of the difficulties encountered
to or beyond their slack. Therefore, the slack of noncritical activities should in implementing software projects within organizations.
also be monitored, to make sure that new critical activities are identified.
Table 6.9. Supply Chain Software Alternative Model Results

!'ftf;::~:~~:<~-t:;.; _
• W:"')·<'." .•
·~
Slack- Slack- Finish date Duration
Form Develop
internal sandbox/ ..
team . . train
....
......
,\.,_,,
Full ERP ven-
dor system
25 weeks 26 weeks 6/7/2013 75 weeks

::.-,:~... '
3-module ven-
dor system
SCM only
19 weeks

9 weeks
19 weeks

10 weeks
12/21/2012

7/20/2012
51 weeks

29 weeks
'i•m•
ilo
In-house 60 weeks 57 weeks 5/17/2013 72 weeks
Open source 24 weeks 18 weeks 8/3/2012 31 weeks
Figure 6.4. Microsoft Project display for partial vendor (I-phase) ASP 12 weeks 0 5/18/2012 20weeks
system.
100 SUPPLY CHAIN INFORMATION TECHNOLOGY SUPPLY CHAIN SOFTWARE INSTALLATION PROJECT MANAGEMENT 101

,
~Jeifi; iG;tJ.~,J>J.lv.
_,;,,>• '. ;. ,· :)>~ ·.. ;tal
~ett ,,\·t ' ', ', <~; '·'~'-}·' k. , '/ ','

U.i~·~~~·~~f~V~.(D~4,liki'~sueh
«>~giuu~• ~;;•.,\;Iwt#,:'fta~ S<),•t·•··~~,,the effi.,.
cien~·<Ji·k~;~,~~.t}\~ugh. aµ,~A1ati9ti.··N~d<\:has ~:irJY~~'lir~~1e:~t:r:
~<~,.~s~,,r~s.U;6~,~~ein. po~~i6~ for .15 ~sef}ltive re ....~iii w'~J 0si~,,.. . train~p,g ~·6
~· ~chi._~.t
''/\0'··.. "·<·:t:J·" ',, .
2.fr>.i'~··~
·.· , ' .. .
~·-·<t;·~}"
~l:iput•t:t~Ho!Jof
.. "; ,, ·,· ,,., ',,
~-¢,..·:·
them. •~n
' .-:
. months 'on th~ n~·Sys~pl, '!~~ die¥~t ~f the" teainicilµlS retel~ect> . . · ' . .· ••,,, "" . • ·. · .. ,A

LaS'V~ 9::~· N~sJ)~,~mpiOyecl;.aboµ~2,2()0 peoplem an f~~~µrf:'~t'.se·,!~is tiilfiing1j<>~tinu~ ~r6uglt~ugu~:!9j~r· ·. ·'·.·. ·

~<()~ ~.Jt999:' res~~-·to .~nfo. m.itites0!Jla~g ~ motor , .· i~nve~t~n ~~~;~~~e.~'9aY'$·~~~ was to ~·none·on !'•;
vekiclda~~t'.im~~'f. ·• ·. ·. '. . . . :.· ·l; \.• ·3-d<\i,lv>lidlj~kend.N t11elau~~ dac,:.ri~rCa/i~wa$.!(Jaxed'qntil ·
fo 199$L~ ·~ hand1- ov! 130~00() vehicle ~tradon
i'; '·• rig'.tttr . . be:'; :; :., ".· · f~b ;3}tTh' J • .···· ..··. /1>7 • >"h :

ren~ed . >
toood;:new·~~
. • } . . . .. . .··· '
....,;;'fi..;M*~• Mtmdnlh. ; .in addition
."'&"'......__... r~ .. .
·~ .·p~m.• ·.t.f'i~ '.'.~ ~."'
~~. ·.·. •......:
;ti,~ty;w~t?Ji~d':hea'r'~~~ ·~.:.'"~.' ~.,f9.,rc1·n·.,
.. . '.: 11,.C;.•..•p····..... ..";. >' ••..·...·•.. ".·:,:· ··.··.u·gs·.··.·.•.··.·. ·. .
~lid~fi ·inter¢St. ~;
g. ·. o. ut . . ·e...
'!;here
to Ej0,Q®ll-~(t6~t®;ne.wdr-l·es·pet month; • stilt''bugs i,r~p-\;et;~ut the·,,ve~~i~red.~n go~Jt,g
Thel~U..~•~1~1i.:to -Mic Wltl ~iceably htcrcasing.
'(he~fO~~:I~rtte;~,s~~~~cequal~.
"l)f }~~:D"" ~..·Pt$~ .~pest$, •n ,,J~6;.a 7f;ear
~~~cit.;; 111<?~.t\0
up~j'of~.~!l~t:t}~$\'.~e~. }i'~~tlesaw<systems
. . . . ,. ~ll~ with\~j~~~nt•fd~~~'~th t1\e ~lodi~g
worldoij, bUt chef were'~ahle to ~ Sbpatate'systems handl- ue·
registratiori ~ Qt.fes-; dli~sJ~;ihfot:mation. This ~ to the .
need for multiph! u~1n6 • ofttti:. ~er inforinatiQll was !tot
current, even though' cit~ ~iUSti~techfaia on•~ne of the sys-
tems. There often . . . ¢Qniplt•J;:ras'1ts, and tt!thni<:fansifwere not
trained to work~ bo$ S)'ltCms, meatimg,that citizens tequiring set- ..
vice in both systems entered a, sefondl~ p soon as their wait in the
first line .was.o¥C£. ~~nots~· . .ucti9:n i~liew hiring a.tl4
intpr9vei~~Fi~·· . ; .. . .• . ag~. ~t ·tiffi~ in . ·Lis Vegas ~· req~ce<i:,;~o ·
Th1t~prqj~t ·~· ~e·~. J:>MV $34 ·.mmiorl;comp~eted hi. .L~ c.q';!l
·e. 't'>itt system.····~JS: · .. n~
.. .. , · " '
tbok 4..,, filtnfltes
. ...
·•.
,,

-
200$.>ihf~li't in~r.~n~·1~nlin~ Septem~ 7,~ 19?9. The
.~~;i-~.-. pnl~i ·fidu¥ ~tem ·~#me~t. eontinu-
~us i-to~~ o~~·nal; lhanSCt and change i'nanaP1~~t.
,is ssn:~ ti~' in~~d f~if5 lffiilufeS,).
~ autlJ~ b ·. a:.Jt1oa:islative a· ·ril'o' rla- ·.
i: . ·. ,Jv·. ". Y ··~;,
·in'I 9~7~.Th~~ in~reased by $8~3 million. iii{\
. ,P.r P '·'

A iletailld: plan 0£'.ia•~ion wai ,ifDet:atCd. arid vendor ptoducts ~f)9~:ft5 :ihillio~ . Jiri ·~n· spen,;~f;Sy ~pv~~er .
· •.\:Tit~~ btti1t on the fitsq>1-e "1th. tbe intent of •i
;claim~ tli,~~hC)"~~ii<l not ~Cd rft'1r budget.
p~dj~ ~rvi~~.out,Utof.Ht•:f ~a core system, HoweWr,· · ~t i~ti<fe i*~ o9riJ~~·
. ·,,, /;
with.~t~ give~ in~~db~.~t dri~er'iiicense, registra- Thisp1;0j . ~ui~;qe~~ l!llpl(t~entation, en"tounterep,,.
tion,lnd Jlf: reti~irements; Aphone-p~cessing center was created to a great deal of ,culty beca-a gttat~of ~iprogram~~.ng ~d
. i., > ' •• ; .'~' • ».', .. v , ',.-,..
102 SUPPLY CHAIN INFORMATION TECHNOLOGY

system installation needed to be accorn~, ~ .... projectl

highly difficult to estimate in terms<>f'.ci.rne {and.d\at i,P.pac$ money)• CHAPTER 7
Typically cost overruns occur, al~~.IT p~.tlel~yfind.;i
way to justify some statement to theeffucttha~tbceyare:W~thin budget.
Recapitulation
Sourc'-.¥apted frpm Dhillon and Cal~~.(2008).

Conclusion This concluding chapter discusses three issues related to current events
in enterprise systems. We want to emphasize the importance of training
Proper planning and management is very important in supply chain
in obtaining effective use of enterprise resource planning (ERP) systems,
management software implementation. Project management in the
implementation of such software includes requirements to accomplish supply chain management (SCM), advance planning systems (APSs), or
electronic information systems (EISs). ERP systems have been around
the following:
long enough and have had sufficient research leading to improved sys-
a. Software and hardware selection tems that upgrades have become impcrtant. This creates great pressure
b. Business process reengineering to gain effective advantage from the on organizations that perhaps spent billions on massive systems in the
new system 1990s, now finding that vendors are suggesting they go through another
c. Assessment of the ERP or SCM readiness of the organization and round of massive installation and retraining with significant capital
plan change management to make the new system effective requirements. Upgrades are encouraged by vendors through dropping ser-
vice on their older systems. (SAP announced around 2004 that they were
Proper project management includes the selection of an effective proj- discontinuing service on their flagship R3 system as of 2007. This was
ect management team. Implementing an EIS is not an inconsequential followed by client outcries, resulting in postponement of discontinued
effort. Considerable attention should be given to the preliminary stages service until 2011.) A final topic is the evolution of added functionality.
of planning. User requirements and selected functionality should be In the past, systems such as APS or customer relationship management
communicated to all involved in the project. Critical success factors need (CRM) were added to ERP systems, with software provided by external
to be considered to carefully coordinate the efforts of the many people vendors. This was changed when Oracle purchased Siebel Systems, the
involved, to include consultants and managers. Gaining employee com- leading CRM vendor around 2005. Oracle thus made CRM a module
mitment to the EIS project is very important. Because the EIS integrates within their ERP. SAP responded by purchasing their own CRM to make
diverse company elements and data from many departments, the effects into a module. The same was done with SCM systems, to include APSs.
of delays ripple with dramatic impact. Therefore, control of the EIS
implementation project is paramount.
Training
In any supply chain software implementation, it is generally understood
that training is a key component of organizational change management
and of the overall success of the implementation. Many important issues
remain in making SCM systems work. User training has shown up as a
critical success factor in the implementation of an ERP in many studies.'
RECAPITULATION 105
104 SUPPLY CHAIN INFORMATION TECHNOLOGY

Managers oft:cn underestimate the magnitude required in such a training • Placing employees in software-specific training, without atten-
program. Training is typically underestimated and is often the first target tion to business processes
• Focusing training on command sequences without explanation
for budget cutting. A period of about 1 year is .usually required until
of why skimping on training time
the trauma of new system implementation passes.This difficult period is
• Continuing tendency of new users to solve problems the old
easier to cope with if a good, thorough training program is adopted. Fur-
way rather than learn the new system
thermore, those organizations that do a poor job of training have been
found to hav-e poorer performing enterprise systems. 2
Good training programs can pay off in many ways. Gartner has Training Media
claimed that each hour of effective training is worth 5 hours to the orga- A number of tools are available to deliver user training. One-way channels
nization because well-trained users take less than a quarter of the time included newsletters, road shows, town meetings, a website, and personal
to reach productive performance levels, they require less assistance from appearances by key leaders to inform employees of what was developing in
help sources~ and they spend less time correcting errors. their IT. Interactive communication included workshops, meetings to deal
Thus SCM training programs focus on transactional training (how with specific issues, conference calls, and collaboration websites. Hands-on
the system works). SCM software companies and implementation teams interaction was also developed, to include sandboxes enabling users to play
are normally very good at delivering training that teaches employees how with the system using simulated data prior to using the real system.
to accomplish transactions in the system. On the other hand, running Some of the reasons training in new EIS systems is difficult include
a business entails much more than simply implementation transactions user diversity, the complexity of the new system, and the variety of train-
within a software program. SCM training programs should spotlight on ing methods available. By their nature, EIS systems are going to radically
change how many people do their jobs. The theory of EIS is to inte-
new b'usiness processes.
grate computer support to all aspects of the business, naturally leading
to user diversity. These people also are busy, especially in coping with the
Training Problems requirements of the new system. Training users in new EIS systems can
be extremely expensive, usually over 10% of total EIS system cost.
EIS software itself is rarely the source of implementation problems. Poor
The need for flexibility in timing and place as well as the need for
training of users is usually the source of most implementation problems.
training in specific functions rather than the comprehensive EIS system
Organizations with higher proportions of new employees may find imple-
have made it important to have flexible training delivery means. This has
mentation of EIS easier. Firms with many employees with many years of
led to an entire industry providing EIS training. Many delivery formats
experience within the firm prior to EIS implementation require greater are available, including the following:
levels of change. Conversely, managerial and professional employees are
often easier to convince of the positive impact of EIS on organizational • Web-based virtual training
effectiveness. Further, the degree of change required within the organi- • Computer-based training
zation can have an impact on EIS installation timing. If the system is • Video courses
implemented too quickly, this may not provide sufficient time for the • Self-study books
organizational climate to change. • Training manuals .
Rarely do EIS implementations run smoothly. The following arc some • Pop-up help screens
of the pitfalls organizations typically face: • Classroom training
106 SUPPLY CHAIN INFORMATION TECHNOLOGY
RECAPITULATION 107

• Sandboxes, or prototype systems using simulated data for

rather it replaces (or upgrades) the old version with a readily available
hands-on orientation
new version from the vendor. However, a lack of experience may cause
the costs and length of the upgrade project to approach or even exceed
Enterprise system training should be conducted during business hours
those of the original EIS/ERP implementation effort. General benefits
to indicate its importance. 3 V. Vathanophas saw three levels of training,
for organizations from EIS/ERP upgrades include the following:
where consultants and vendors can first train IT staff, who in turn train
individual departmental representatives (sometimes called superusers),
• Eligibility for help desk support-Most of software vendors stop
who finally deliver training to their compatriots within their departments.
providing technical support 12 to 18 months after the next
The scope of training is demonstrated by the experiences of Pratt
version becomes available. Therefore, keeping up with the pace
and Whitney, Canada (P&WC) in 1998. 4 They trained 110 employees
of vendors will guarantee the support for the system from the
from their six most affected departments as internal trainers (superusers). vendors.
The year 1998 saw P&WC convert their facilities into a massive class-
• Solutions for outstanding bugs or design weaknesses-It is impos-
room, training over 3,000 employees in both technical aspects (system
sible to guarantee spotless and error-free systems after the
navigation and task training) delivered by consultants as well as providing
implementations even though vendors will conduct many dif-
business-oriented training (processes and tasks). P&WC created over 150
ferent testing processes to eliminate the occurrence of errors in
manuals to cover diverse user needs.
the system before the leasing time.
• New, expanded, or improvedfeatures-Software provides
Upgrades organizations the knowledge and strength (i.e., best practices)
from the vendors. Upgrades provide organizations future
Upgrades are mainly intended to take advantage of new technologies and
enhancement from the vendors to give the organizations better
business strategies to ensure that the organization keeps up with the latest
opportunities to catch up the current business development,
business development trends. Therefore, the decision to upgrade SCM
improve their processes and build more efficient business mod-
and related systems is usually not driven by code deterioration or antici-
els with new functions, new features and new processing styles
pated reduction in maintenance costs alone but by different purposes.
provided in the upgraded versions.
According to an AMR study, 55% of ERP upgrades were voluntary
business improvements triggered by the need for new functionality,
expansion, or consolidation of systems; 24% of upgrades were triggered Add-ons
by technology stack changes; 15% of upgrades were forced by discontin-
Add-on (or bolt-on) is ERP jargon for third-party applications. More
ued support of the running version of software to avoid vendor support
specifically, an add-on is an execution system providing very specific
termination; and 6% of upgrades were triggered by bug fixes or statutory
functionality or technology to complement ERP software. Many useful
changes._
applications of this type are available. The types of software and related
The cost of upgrades is high. Upgrade costs may involve 50% of the
features of add-on software listed by Microsoft for their ERP software are
original software license fee and 20% of the original implementation cost shown in Table 7.1.
per user, which means over $6 million for a 5,000-user system. Typi-
Table 7.1 shows a variety of functions that can still be supported by
cally, each ERP upgrade requires 8 to 9 months of effort with a team the
add-ons. In the 1990s, major functions were supported, such as SCM
equivalent of one full-time employee per 35 business users. The adopting
and CRM. The evolution in ERP systems has seen these major functions
organization does not have to develop and rewrite the system itself, but
pulled into more integrated ERP systems. However, there will always be
108 SUPPLY Cl IAIN INFOR~1ATION TECHNOLOGY
Ill ! 111

Table 7 .1. Microsoft ERP Add·ons ~

Add-on function Vendor
Mortgagt· Bank lmplcn1t:111.1 1ti~•I! o l
Au tom11t1ve ERP AIM Computer Solutinns, Inc.
Enterpri~c Applkation l111t·~r·11 1i,1n
Disrrihunon, finanu.11, e-commerce Atha Spectrum
l larmond Bank served the consu mer b.111ki11~ 11111l.:c1 111 l1d111HI,
111tq.:ra11on
Home loans were among irs most proficablc bu11rnt·\s .1l·1ivi1ies. 'Ilic
Sales anJ use rax Av:1 lara
bank emphasized customer service and sought co lower cost~ duou~lr
E-commerce Azox
use of e-business. Much of their mortgage loan process was conducted
AuwmarcJ EDI , XML pn><;c,smg Darn M,151ms Sufrware
Enc11rc Business Solutions
over the I ncernec. 1 lowever, development o f diverse systems over time
Project tracking, hn.rncial
led to many nonintegraced systems. Problems were encountered in
Electron1~ mvmcmg Enliven Sofrwarc
efforts to incegrace chese systems:
Support Excel tr:insl,1tion tll Dynamics lll' lnlin1a flusiness Technology
P<
1rroll ,mJ hum.in R"'<>UtCCS lnregrit\' Data
Papcrlc" ERP support Mc1.1fl le Inform<1t11in Systems • T he bank's web portal wirh online forms for application
flar code J,1ta c:iprure Panatrnck, Inc. was based on HTML and Java, running on a Unix-based
PapcrS.wePm
Apache Web server.
Document 1111.11:mg
Project resource managcmcn r Tcnrnx • A back-end mortgage application package ran on Windows
Batch m.mufacturulJ! support Vicini!\' .\fanuf.1ctur111g NT, obtained from a fi nancial services IT fi rm.
• A legacy account management system ran on a mainframe
ideas to supplement ERP systems, some of which will prove commercially using a VME operating system.
viable. A snapshot of some of those syscems is d emonstrated by Table 7.1. • A CRM package ran on Windows NT.
Supporr of customer relationship management is the form of data min- • A Siebel call-center application rn n on Windows NT.
ing most commonly associated wirh ERP. CRM allows businesses co identify • Other software accomplished credit checking, address rar-
the prohrabiliry of specific cuscomers, and to increase chances of retain- ing, and ocher mortgage check.
ing chem. This is accomplished by having all relevant information readily
available that is needed for planning, product, and service chroughouc the The lack of imegration meant char the ban k reemered data across
customer life cycle. SAP has been a leader in enhancing their product's abili- systems, introducing inaccu racies and slowing the loan process. CRM
ties co support CRM. 6 Many of these systems fuiled, as they introduced the was difficu lt to implement, because while some syscems in che bank
need for middfeware, or software that translaces daca from one vendor sys- had information about specific customers, chis information could nor
cem to another, creating an added layer of complexiry and expense. be accessed by cross-selling staff.
Nacurally, ERP vendors added functionaliry such as SCM and CRM A team of 12 business and IT professionals was formed within the
to more efficiently utilize their systems. As we have mentioned, char is bank and given 24 months ro integrate the banks information sys-
whac Oracle did when they purchased Siebel Systems' CRM product in tems. Four external consulcams were hired to give the ream assistance.
Sepcember 2005 and integrated it within their ERP (ch us creating an T he first 12 months were ded icaced to the home loans divisions. Of
EIS in ind ustry jargon). T hus add-ons became modules th rough acquisi- chis year, 3 months were devoted to BPR, mapping existing morcgage
tion. C lients benefi ced by elimi nation of rhe need for middleware for that processes and improving chem. This led to a plan th<it would reduce
application. However, realiscically, there will always be add-on products curnaround rime in morcgage processing by 50% and would seam-
of some type generaced by che accive sofrw,rre developmenr industry. lessly incegrace and transfer information across bank systems. The plan
110 SUPPLY CHAIN INFORMATION TECHNOLOGY RECAPITULATION 111

would be scalable from the current 120 mortgage applications per avoid duplication and to control updates. System integra-
week t9,over 300, with acceptable security. tion solved this, but initially it was a challenge to determine
Enterprise applicatioh · integration (EAl) is a system consisting which systems used which data.
of mlddleWare to enable communication acros$ applications, adapt• • It was found to be more sensible to replace some applica-
ers to conneet these applications to the overall infrastructure, and tions rather than integrating them. The bank's old CRM
an in,tegration broker 'capturil:lg business rules controlling workflow application was dropped. replaced by a new Siebel call-
and messaging. £AI significantly reduces the number of. interfaces center application.· Using software from as few sources as
that might be needed, and all components woulq link to the broker possible had value.
comp9nent. Ve~dors providing EAi ·tools include IBM, TIBCO, • EAi requires considerable testing.
Wel:,~ethot{s, SeeBeyond, and Vitria. • Phasedrolfout is preferable, controlling the risk of bringing
The consultantS suggested anEAi for the bank. with a new Siebel- systems online.
ba$ed CRM system· more closely coupling CR.M with. the call center
Source: Adapted from Lam (2005).
operation. A new plan to insraJI the EIA system in three phases was
proposed qver a year. Testing led to identification of bugs, ex.tend-
ing the proJectduration 3 months beyond what had been planned. Conclusion
The original 1~..monthplan was sciU not cornpletedin18 months, . Training is a key component of a successful enterprise system installation.
and project costs exceeded the original budget by 60%, However, the Training needs to be considered in the initial project budget. Typically,
bank's sys~erns weremuch more efficiently integrated. it is underestimated by significant amounts. There are two major ele-
The bankconsidered the effort to have been well-worthwhile, lead-
ments of enterprise system training. The first is focused on how to use
ing to a number of key lessons learned:
the system, and this type of training is well-developed by vendors and
consultants. The second is on organization specific business processes.
• It is. best to understan4 business processes before im ~le­
This second form of training has proven to be far more important than
menting systems, involving all stakeholders involved.
• Selection of integratfon architecture is important consider- the first. Vendors and consultants can't be expected to deliver training
ing the number of applicationsto be integrated and the programs covering organization-specific processes unless the installed
need for real~time4ata transfer. EAi is not always the appro- system has no customization. Usually, effective systems that match orga-
priate choice. nizational needs do have customization, and the organization itself will
• EAi packages sel~cted need to have whatever adapters are have to develop chis form of training. (They will want to if it covers core
required forthe software to be used within the system. competencies chat yield competitive advantage to the organization.)
Ad,apters typically costaroun4 $50,000. An effective means to organize training is to have experts {vendors or
• Some (;UStom cooing will be required to linkadapters. In consultants) train IT staff. IT staff in turn train a set of superusers from
the bank's case, the Siebel adapter could be configured using departments within the organization, who then relay the knowledge they
~ GUI interface without coding. But the fllOrtgage applica- obtain to general organizational users.
tion.pacl<.ige Was Java-based, .and the adapter n,eeded to be There are many different media available to deliver enterprise system
coded., which proved difficult. training. One-way media can be used to inform users of the system's value
• The organizations data structure needed to be an11;lyzed to
to the organization. Two-way media are usually more effective in teaching
112 SUPPLY CHAIN INFORMATION TECHNOLOGY

users how to use the system. Hands-on interaction with sandbox systems
can be highly effective in training users of their specific job requirements.
Through care in the planning and delivery of training, the success rate Notes
of enterprise systems installations can be vastly improved.
EIS/ERP upgrade projects have grown in importance, as vendors are
seeking to generate revenue through improved systems. The reticence of Chapter 1
vendors to support old systems was noted by multiple organizations in
1. Ptak and Smith (2011).
this study. (The value of improved functionality was also noted.)
2. Manetti (2001).
Upgrade projects seem to be much more controllable than initial
3. Retrieved July 11, 2011, from http://www.advanced-planning.eu/advanced
supply chain management software installation projects. This should planninge-3 56.htm
be expected due to the experience organizations gain with their origi- 4. Moser and Ward (2008).
nal systems. All the organizations seemed to do something that fit the 5. Gonzalez (2007).
theoretical model of an upgrade project that we used. Assessment, plan- 6. Saenz de Ugarte et al. (2009).
7. Olson and Kesharwani (2010).
ning, and action phases were present to at least some degree. The renewal
8. Olson (2004).
phase noted by the fifteen organizations involved very smooth turnover.
9. Mabert et al. (2000); Olhager and Selldin (2003); Katerattanakul et al.
A limitation of the study is that future implications were not yet available (2006).
in all cases (problems may crop up later), although all organizations cred-
ited strong planning and project management as ways to assure smooth
Chapter 2
transitions.
The software industry continues to generate new applications, and 1. Fishman (2006).
improvements on existing applications. That is progress. Large vendors of 2. Holzner (2006).
3. Brady et al. (2001).
software will add such systems (or develop their own), leading to the need
4. Mabert et al. (2000); Olhager and Selldin (2003).
for upgrades and renewing their revenue stream. Smaller vendors such as 5. Stevens (2001).
Sage and Lawson may not add as many of these functionalities, so that 6. Davenport (1998).
there will be an increasing variety of supply chain software varieties avail- 7. Davenport (1998).
able, calling for more complicated software selection decisions, to include 8. Stadtler (2005)
open-source options. The price of progress is often more complexity. 9. Wiers (2009).

Chapter 3
1. Schwartz (2003).
2. "Outsourcing's next generation" (2003).

Chapter 4
1. Hall et al. (1993).
2. Scott and Kaindle (2000).
3. Hammer and Stanton (1999).