Enterprise architectural framework for supply‐chain integration

Enterprise architectural framework for supply-chainintegration

Charu ChandraIndustrial and Manufacturing Systems Engineering, University of Michigan,Dearborn, Michigan, USASameer KumarPrograms in Manufacturing Systems & Engineering, University of St Thomas,St Paul, Minnesota, USA

Introduction

A supply chain is a society (a network of

members, termed a group) formed by

autonomous entities (and their systems) by

bonding together to solve a common problem.

With their collective and collaborative

efforts, they sustain the progress of each

member as well as the group. Collaboration

between members requires effective

communication. In a collaborative

environment, a member may modify its

norms of behavior to accommodate other

member's perspectives (Bond and Gasser,

1988; Gasser, 1991; Moulin and

Chaib-Draa, 1996).

This paper describes an enterprise

architectural framework with an example of

a generic supply chain from the textile

industry. Its focus is on enterprise

integration issues, such as in a supply chain.

It is organized as follows. First, an historical

perspective on supply chains aimed at

highlighting significant enterprise

integration issues is presented. Designing

and modeling solutions for these integration

issues is the motivation for this research,

which led to the framework. Finally,

implications of this research on designing

and modeling complex supply chains

generically are noted.

An historical perspective on supplychains

The significance of changes taking place in

supply-chain initiatives can be best

appreciated from a review of historical

aspects of production and operations

management activities (Poirier and Reiter,

1996).

During the period from 1960 to 1975,

corporations had vertical organization

structures and optimization of activities was

focused on functions. Relationships with

vendors were win-lose interaction, many

times adversarial. Manufacturing systems

were focused on materials requirements

planning (MRP).

From 1975 to 1990, corporations were still

vertically aligned but several were involved

in process mapping and analysis to evaluate

their operations. Organizations were

beginning to realize the benefits of the

integration of functions such as, product

design and manufacturing. Various quality

initiatives, such as the total quality

management philosophies of Deming, Juran,

and Crosby; and ISO standards for quality

measurement were initiated by many

organizations. Manufacturing systems were

focused on MRP II.

During the 1990s, corporations experienced

increasing global competition. Strategic

alliances among organizations have been

steadily growing. Organization structures

are starting to align with processes.

Manufacturing systems in organizations

have been enhanced with information

technology tools such as, enterprise resource

planning, distribution requirements

planning, electronic commerce, product data

management, and collaborative engineering,

etc. (Aberdeen Group, 1996). Design for

disassembly, synchronous manufacturing,

and agile manufacturing are some of the new

paradigms in manufacturing. There has been

a growing appreciation in many firms of total

cost focus for a product from its source to

consumption, as opposed to extracting lowest

price from immediate vendor(s) (Turbide,

1997). There has also been an increased

reliance on purchased materials and outside

processing with a simultaneous reduction in

the number of suppliers and greater sharing

of information between vendors and

customers. A noticeable shift has taken place

in the marketplace from mass production to

customized products. This has resulted in the

emphasis on greater organizational and

The current issue and full text archive of this journal is available at

http://www.emerald-library.com/ft

[ 290 ]

Industrial Management &Data Systems101/6 [2001] 290±303

# MCB University Press[ISSN 0263-5577]

KeywordsSupply chain,

Supply-chain management,

Enterprise economics

AbstractThe concept of supply chain is

about managing coordinated

information and material flows,

plant operations, and logistics. It

provides flexibility and agility in

responding to consumer demand

shifts without cost overlays in

resource utilization. The

fundamental premise of this

philosophy is; synchronization

among multiple autonomous

business entities represented in it.

That is, improved coordination

within and between various

supply-chain members. Increased

coordination can lead to reduction

in lead times and costs, alignment

of interdependent decision-making

processes, and improvement in

the overall performance of each

member as well as the supply

chain. Describes architecture to

create the appropriate structure,

install proper controls, and

implement principles of

optimization to synchronize the

supply chain. A supply-chain model

based on a collaborative system

approach is illustrated utilizing the

example of the textile industry.

process flexibility and coordination of

processes across many sites. More and more

organizations are promoting employee

empowerment and the need for rules-based,

real-time decision support systems to attain

organizational and process flexibility, as well

as to respond to competitive pressure to

introduce new products more quickly,

cheaply and of improved quality.

The underlying philosophy of managing

supply chains has evolved to respond to these

changing business trends. Supply-chain

management phenomenon has received the

attention of researchers and practitioners in

various topics. In the earlier years, the

emphasis was on materials planning

utilizing materials requirements planning

techniques, inventory logistics management

with one warehouse multi-retailer

distribution system, and push and pull

operation techniques for production systems.

In the last few years, however, there has been

a renewed interest in designing and

implementing integrated systems, such as

enterprise resource planning, multi-echelon

inventory, and synchronous-flow

manufacturing, respectively. A number of

factors have contributed to this shift. First,

there has been a realization that better

planning and management of complex

interrelated systems, such as materials

planning, inventory management, capacity

planning, logistics, and production systems

will lead to overall improvement in

enterprise productivity. Second, advances in

information and communication

technologies complemented by sophisticated

decision support systems enable the

designing, implementing and controlling of

the strategic and tactical strategies essential

to delivery of integrated systems.

In the next section, a framework that offers

an unified approach to dealing with

enterprise related problems is presented.

A framework for analysis ofenterprise integration issues

As mentioned in the preceding section, the

availability of advanced production and

logistics management systems has the

potential of fundamentally influencing

enterprise integration issues. The motivation

in pursuing research issues described in this

paper is to propose a framework that enables

dealing with these effectively. The approach

suggested in this paper utilizing supply-chain

philosophy for enterprise integration

proposes domain independent problem

solving and modeling, and domain dependent

analysis and implementation. The purpose of

the approach is to ascertain characteristics of

the problem independent of the specific

problem environment. Consequently, the

approach delivers solution(s) or the solution

method that are intrinsic to the problem and

not its environment. Analysis methods help

to understand characteristics of the solution

methodology, as well as providing specific

guarantees of effectiveness. Invariably,

insights gained from these analyses can be

used to develop effective problem solving

tools and techniques for complex enterprise

integration problems.

The discussion of the framework is

organized as follows. First, the key guiding

principles of the proposed framework on

which a supply chain ought to be built are

outlined. Then, a cooperative supply-chain

(CSC) system is described as a special class of

a supply-chain network implementation.

Next, discussion on a distributed problem-

solving strategy that could be employed in

integrating this type of system is presented.

Following this, key components of a CSC

system are described. Finally, insights on

modeling a CSC system are offered. Key

modeling principles are elaborated through

two distinct modeling approaches in the

management science discipline.

Supply chain guiding principles

Firms have increasingly been adopting

enterprise/supply-chain management

techniques in order to deal with integration

issues. To focus on these integration efforts,

the following guiding principles for the

supply-chain framework are proposed. These

principles encapsulate trends in production

and logistics management that a supply-

chain arrangement may be designed to

capture.. Supply chain is a cooperative system. The

supply-chain arrangement exists on

cooperation among its members.

Cooperation occurs in many forms, such

as sharing common objectives and goals

for the group entity; utilizing joint

policies, for instance in marketing and

production; setting up common budgets,

cost and price structures; and identifying

commitments on capacity, production

plans, etc.. Supply chain exists on the group dynamics

of its members. The existence of a supply

chain is dependent on the interaction

among its members. This interaction

occurs in the form of exchange of

information with regard to input, output,

functions and controls, such as objectives

and goals, and policies. By analyzing this

[ 291 ]

Charu Chandra andSameer KumarEnterprise architecturalframework for supply-chainintegration


information, members of a supply chain

may choose to modify their behavior

attuned with group expectations.. Negotiation and compromise are norms of

operation in a supply chain. In order to

realize goals and objectives of the group,

members negotiate on commitments made

to one another for price, capacity,

production plans, etc. These negotiations

often lead to compromises by one or many

members on these issues, leading up to

realization of sub-optimal goals and

objectives by members.. Supply-chain system solutions are Pareto-

optimal (satisficing), not optimizing.

Supply-chain problems similar to many

real world applications involve several

objective functions of its members

simultaneously. In all such applications, it

is extremely rare to have one feasible

solution that simultaneously optimizes all

of the objective functions. Typically,

optimizing one of the objective functions

has the effect of moving another objective

function away from its most desirable

value. These are the usual conflicts among

the objective functions in the multi-

objective models. As a multi-objective

problem, the supply-chain model produces

non-dominated or Pareto-optimal

solutions. That is, solutions for a supply-

chain problem do not leave any member

worse-off at the expense of another.. Integration in supply chain is achieved

through synchronization. Integration

across the supply chain is achieved

through synchronization of activities at

the member entity and aggregating its

impact through process, function,

business, and on to enterprise levels,

either at the member entity or the group

entity. Thus, by synchronization of

supply-chain components, existing

bottlenecks in the system are eliminated,

while future ones are prevented from

occurring.

A cooperative supply-chain

A supply-chain network depicted in Figure 1

can be a complex web of systems,

sub-systems, operations, activities, and their

relationships to one another, belonging to its

various members namely, suppliers,

carriers, manufacturing plants, distribution

centers, retailers, and consumers. The

design, modeling and implementation of such

a system, therefore, can be difficult, unless

various parts of it are cohesively tied to the

whole.

The concept of a supply-chain is about

managing coordinated information and

material flows, plant operations, and

logistics through a common set of principles,

strategies, policies, and performance metrics

throughout its developmental life cycle (Lee

and Billington, 1993). It provides flexibility

and agility in responding to consumer

demand shifts with minimum cost overlays

in resource utilization. The fundamental

premise of this philosophy is

synchronization among multiple

autonomous entities represented in it. That

is, improved coordination within and

between various supply-chain members.

Coordination is achieved within the

framework of commitments made by

members to each other. Members negotiate

and compromise in a spirit of cooperation in

order to meet these commitments. Hence, the

label(CSC). Increased coordination can lead

to reduction in lead times and costs,

alignment of interdependent decision-

making processes, and improvement in the

overall performance of each member, as well

as the supply-chain (group) (Chandra, 1997;

Poirier, 1999; Tzafastas and Kapsiotis, 1994).

A generic textile supply chain has for its

primary raw material vendors, cotton growers

and/or chemical suppliers, depending upon

whether the end product is cotton, polyester or

some combination of cotton and polyester

garment. Secondary raw material vendors are

suppliers of accessories such as, zippers,

buttons, thread, garment tags, etc. Other tier

suppliers in the complete pipeline are: fiber

manufacturers for producing the polyester or

cotton fiber yarn; textile manufacturers for

weaving and dying yarn into colored textile

fabric; an apparel maker for cutting, sewing

and packing the garment; a distribution center

for merchandising the garment; and a retailer

selling the brand name garment to consumers

at a shopping mall or center.

Synchronization of the textile supply chain

is achieved through coordination primarily

of:. replenishment schedules that have been

passed on through the echelon from

retailer onwards to apparel maker, textile

manufacturer and fiber manufacturer;. commitments made on capacity

utilization between echelons, such as

apparel maker and textile manufacturer

and so on; and. commitments made on product costing at

various stages, that is raw materials,

fiber, textile, garment between various

members in the textile supply-chain.

Various textile supply-chain members enter

into negotiation with one another and

[ 292 ]



eventually compromise on such issues as

reserving manufacturing capacities, and

holding consumer price levels through cost

sharing.

The common metric for a textile supply

chain is removing waste through highly

coordinated decisions, in order to reduce

lead-time and inventory levels at various

stages of the product life cycle.

A distributed problem-solvinghypothesis for a cooperativesupply-chain system

A classic problem encountered by a generic

textile supply chain is that of planning and

coordinating supply-chain production to

meet consumer demand, while making

effective use of resources and promoting

cooperation among members, so as to achieve

lead-time (waste) reduction.

Building blocks of an integrated

framework to manage a supply chain such as

the generic textile industry example are ±

availability, supply and demand for

inventory; and a planning unit that

coordinates marketing and production of

inventory. Relationships between these are

expressed in traditional `̀ Inventory Balance''

equation, as follows:

Inventory level� Supply level � Demand level

The system architecture of a CSC is based on

the distributed problem-solving approach,

illustrated in Figure 2. The CSC is comprised

of a group and more than one member. The

supply chain is arranged in the order the

flow of materials, processes, and information

occurs between its members. In the textile

industry example depicted in Figure 1,

consumer demand is relayed by retailer to

the apparel maker, textile manufacturer,

fiber manufacturer, and ultimately to cotton

grower. Similarly, flow of material occurs in

the transforming of cotton to yarn by the

fiber manufacturer, fabric by the textile

manufacturer, apparel by the apparel maker,

and finally a name brand garment by the

retailer. The interaction between members

occurs as a consumer and a provider. Thus,

an apparel maker assumes the role of a

provider (of apparels) in its dealings with a

retailer (a consumer of apparels). However, it

acts as a consumer of fabric while dealing

with a textile manufacturer (a provider of

fabric). The CSC requires design and

implementation of three primary

components: structure, control, and

optimization. These are described below.

StructureThe CSC is a distributed system of collection

of components and elements of autonomous

business entities. In the distributed problem-

solving environment, the task of solving a

problem is divided among a number of

modules or nodes (autonomous entities and

their systems). They cooperatively

decompose and share knowledge on the

Figure 1A supply chain network

[ 293 ]



problem and its evolving solutions.

Interactions between members in the form of

cooperation and coordination are

incorporated as problem-solving strategies

for the system. Entity group is responsible

for coordination throughout the supply

chain. Entity member brings specialized

expert knowledge and technology to the

supply chain. The decision-making process is

centralized for the group. The group enforces

common goals and policies of the supply

chain on its members. However, decision-

making at member level is decentralized.

Each member pursues its own goals,

objectives, and policies, independently of the

group.

A common knowledge base supports the CSC

structure. Knowledge is assimilated for an

activity (the lowest level of information) in a

specific domain and aggregated for various

decision-making levels in the enterprise.

Building blocks for the generic textile

supply chain for this research are being

modeled after the integrated production

planning and control (IPPC) problem,

whereby each block solves its uniquely

focused forecasting and inventory

management, capacity planning, and

production management problems, but their

actions are coordinated closely through

feedback and feed forward mechanisms.

Each member of the textile supply chain

possesses specialized knowledge about the

textile business. For example, a fiber

manufacturer possesses specialized

knowledge and expertise in polymerization, a

textile fabric manufacturer on jet dyeing, and

a retailer on mail order merchandising.

At the group level for a generic textile

supply chain, common goals are lead-time

reduction and capacity sharing among

members. At the member level, a textile

fabric manufacturer may set up its own goals

of batching, and lot sizing; and an apparel

manufacturer may set up higher inventory

turns to liquidate finished goods inventory.

System analysisThe analysis of a CSC system focuses on

interactions between its design components.

That is, analyzing interactions between

members of the supply chain for information

sharing, defining controls and specifying

roles and responsibilities, while adding value

to the product. This requires identifying

activities in the product life cycle, defining

strategies for their implementation and

establishing performance criteria to measure

their outcome. Figure 3 illustrates a

structured approach to integrate CSC design

components over the product and process life

cycles.

For each member, primary and support

activities in the value chain of a product are

identified. These activities add value to the

product as it traverses its natural life cycle.

Primary activities represent key milestones

in the evolution of the product from its

conception to its natural succession by newer

and improved product(s). In this process,

Figure 2A supply-chain enterprise decomposition model maps a distributed problem-solving architecture

[ 294 ]



primary activities follow natural product and

process life cycles, such as marketing and

research and development activities during

the product and process design phase;

handling inbound logistics of inventory

management, production planning, and

purchasing; operations for materials

transformation to an end-product; outbound

logistics of warehousing and shipping; and

finally after sales service and customer

support. As their name suggests, support

activities are designed to provide support for

primary activities. Similarly, performance

measures are identified for the member,

depending upon the role it plays, that is a

consumer or provider of goods or services, in

the supply chain. Table I, provides a

summary of the output representing a

generic textile supply chain, derived from

analyses utilizing the approach depicted in

Figure 3. As may be noted, as a support

activity, information management plays a

pivotal role in coordinating all primary

activities in the textile supply chain by

gathering, analyzing, and disseminating

information to its members.

For the group, support activities assume a

key role in planning, coordinating, and

communicating pertinent information by

goals, objectives, and policies of the supply

chain to its members as they pursue

execution of primary activities. Performance

of the entire supply chain is measured by key

measures jointly agreed to by members of the

supply chain. Table II, provides a summary

of the output at the group level, representing

a generic textile supply chain, derived from

analyses utilizing the approach depicted in

Figure 3. As may be noted, as a support

activity, information management plays a

pivotal role in coordinating all primary

activities in the textile supply chain through

sharing of information pertaining to key

decision-making variables across primary

activities.

ControlControl in the CSC is maintained via goals,

policies, and objectives that are synchronized

along the system's decision-making

hierarchy. This is accomplished by applying

principles of complementarity, consistency,

and constriction to these control elements, as

depicted in Figure 4. A vertical arrow

between two decision-making levels signifies

complementarity of controls at these levels.

Thus, a primary goal at the strategic level

must be complementary to a secondary goal

at tactical and tertiary goal at the operational

levels. A horizontal arrow signifies

consistency between control elements across

a decision-making level. Thus, a strategic

goal must be consistent with policies and

objectives outlined for its implementation. A

diagonal arrow ( ) denotes constriction

between goals, policies and objectives

Figure 3A supply-chain enterprise analysis approach

[ 295 ]



between decision-making levels. Thus, a

strategic goal will constrain policies to be

implemented at the tactical level, which in

turn will constrain objectives at the

operational level.

The above control hierarchy can be

explained through the example of a generic

textile supply chain.

Table III provides a summary of close

dependence between goals, policies, and

Table IMember enterprise value analysis

ProcurementTechnologydevelopment

Informationmanagement Others

Marketing and sales Buy advertisingcampaignsBuy salespromotions

Consumer marketresearchIncorporate marketneeds in the product

Forecast demandand salesSales analysisTrack productperformance

Coordinate orderprocessing

Inbound logistics(receiving,warehousing,inventory control,production planning)

Procure end-productsProcure rawmaterials forassembly andpackaging

Receive and trackraw materials andend-products

Manage storage ofraw materials andend-products

Plant operations(manufacturing,inspection, productassembly, productpackaging)

Quality inspection offinished productAssemble end-productPackage end-product

Outbound logistics(warehousing,inventory control,shipping)

Procure shipmentmodes

Inventory control offinished productTrack and reportshipments

Select shipment androuting modesConsolidate order fora carrier

Service(organization andmanagement)

Manage inventorycarrying, quality,back order, andopportunity costsAnalyze costvariance

Guarantee shipmentschedules

Table IIGroup enterprise value analysis

Commitments Information management

Marketing and sales Price agreements Share forecast demand and salesShare product performanceShare product cost data

Inbound logistics(receiving, warehousing, inventorycontrol, production planning)

Agreement on inventory stock levelsPre-commitments on short andlong-term manufacturing capacity

Share production forecasts and plansShare production schedulesShare inventory statusReserve manufacturing capacity forspecific products

Plant operations (manufacturing,inspection, product assembly,product packaging)

Pre-shipment inspections Share product and processspecifications

Outbound logistics (warehousing,inventory control, shipping)

Warehousing agreements on finishedgoodsDirect shipments from manufacturinglocations

Share inventory statusShare customer order information

Service (organization andmanagement)

Guarantee delivery schedules Share forecast demandShare forecast production schedules

[ 296 ]



objectives for a member. For example, for a

retailer, a marketing goal of `̀ achieving a

95 percent order-fill-rate of within two days of

order processing'', is a primary goal of

realizing excellent customer service. This

primary goal of the retailer drives the

secondary goal of planning for procurement

`̀ achieving six inventory turns during the

planning cycle'', so as to realize quick

turnaround of inventories, in order to meet a

very high customer service level. The above

primary and secondary goals, drive the

tertiary goal of `̀ achieving a 98 percent

shipment-fill-rate of within eight hours of

order receipt'', so that shipments keep pace

with the inventory management activity

represented by the secondary goal, and fulfill

high service levels imposed by the primary

goal.

Achieving marketing's primary goal of a

high customer service level, and

warehousing operations objective of

minimizing merchandising cost, requires

close coordination between marketing and

operations function in order to implement a

just-in-time (JIT) procurement policy. Thus,

marketing and operations functions may

devise joint policies that enable such

coordination.


high customer service level, may require

implementing a procure-to-stock policy so as

to maintain required safety stocks, and

achieve the objective of maximizing

customer service and thereby increasing

revenue for the retailer.

Table IV provides a summary of close

dependence between goals, policies, and

objectives for a group. For example, for the

textile industry supply chain, a marketing

goal of `̀ achieving an industry benchmark of

95 percent order-fill-rate of within two days of

order processing'', is a primary goal of

realizing excellent customer service. This

primary marketing goal, drives the

secondary goals of production planning

`̀ achieving a 90 percent effective capacity

utilization '', and `̀ achieving inventory turns

of five or above'', so as to:. ensure high utilization of combined

capacities in the supply chain; and. realize quick turnaround of inventories in

order to meet a very high customer

service level.

The above primary and secondary goals,

drive the tertiary goals of `̀ achieving over

80 percent actual capacity utilization'', and

`̀ achieving less than 5 percent rejects'', so

Figure 4A supply-chain enterprise hierarchy of controls

Table IIIMember enterprise hierarchy of controls

Objective(s) Policy/policies Goal(s)

Marketing Maximize customerservice

Implement a procure-to-stock policy

Achieve a x% order-fill-rate of withint days of order processing

Procurement planning Maximize inventoryturns

Implement a JITprocurement policy

Achieve k inventory turns

Warehouse operations Minimizemerchandisingcosts

Implement a quickresponse shipment policy

Achieve a s% shipment-fillrate of within t hours oforder receipt

[ 297 ]



that capacities are utilized as planned, and

rejections are kept to a minimum, thereby

not adversely impacting high capacity

utilization and inventory turnaround

represented by the secondary goal, and

fulfilling high service levels imposed by the

primary goal.


high customer service level, and production

planning function objective of minimizing

manufacturing costs, requires close

coordination between marketing and plant

operations functions in order to implement a

just-in-time (JIT) scheduling. Thus, marketing

and plant operations functions may devise

joint policies that enable such coordination.


high customer service level, may require

implementing a push or pull policy so as to

maintain required safety stocks, and achieve

the objective of maximizing customer

service, and thereby increasing revenue for

the supply chain.

OptimizationThe principle of optimization of the CSC

system is enunciated by investigating

relationships between methods, standards,

and costs on the operations of the enterprise.

Figure 5 depicts the following relationships,

which are described below in the context of a

textile industry supply chain example:. Standards versus costs. The design and

implementation of standards entails costs.

For a large (more than 22,000 members in

the USA), highly complex, and advanced

industry such as textiles, it is especially

true. First, designing standards for a

multi-level industry hierarchy (retailer,

apparel maker, textile manufacturer, fiber

manufacturer, cotton grower) with

diverse membership (from small family

businesses to large corporations) is a

logistics challenge for a designer to solicit

feedback from participants. Further, and

more realistically, it is a technological

challenge because of the highly diverse

levels of technology (from manual to

highly automated plant operations) in use

throughout the industry. Second, after the

standards have been designed, their

implementation through practical

demonstrations and training is cost

prohibitive and time consuming,

considering the size of the industry.. Standards versus productivity. Standards

are implemented to bring uniformity to

established and often repeated tasks.

Standards rolled out as procedures and

corresponding time and costs are used as

yardsticks to measure productivity. In the

case of a textile supply chain, an analysis

challenge will be as how to interpret

productivity when the level of technology

and associated tools and techniques is

diverse among its membership.. Standards versus methods. Standards,

when proven, are classified as methods.

Methods, in turn may influence operating

practices and policies of the firm.. Influence of methods and standards on

product and process designs. Standards

and methods can influence product and

process design, especially when product

characteristics change significantly as to

affect process characteristics and warrant

design modifications. In the case of ta

extile supply chain, product and process

designs are impacted when coloring,

dyeing, and fiber weight and density

constraints are imposed on the raw

materials.

These relationships are first quantified

through known work design and methods

improvement techniques and then

represented as joint attributes of various

decision-making levels and control elements

in the CSC system model (Nadler, 1970;

Niebel, 1993).

Table IVGroup enterprise hierarchy of controls

Objective(s) Policy/policies Goals(s)

Marketing Maximize customerservice

Evaluate and implement apush or pull policy

Achieve an industrybenchmark of x % order-fill-rate of within t days oforder processing

Production planning Maximize productionunder-runsMaximize inventory turns

Evaluate and implement aJIT scheduling or plannedproduction schedulingpolicy

Achieve a y % effectivecapacity utilizationAchieve inventory turns ofk or above

Plant operations Minimize manufacturingcostsMaximize yield perproduction run

Evaluate and implement aJIT manufacturing orplanned manufacturingpolicy

Achieve over z % actualcapacity utilizationAchieve less than r %rejects

[ 298 ]



Cooperative supply-chain system:a member perspective

Members of a CSC system in a distributed

problem-solving architecture are

heterogeneous (Durfee et al., 1989). The

degree of heterogeneity can be attributed to

the implicit and explicit behaviors they

portray within their internal organization, as

well as in their dealings with other members

of the supply chain. Some key factors that

differentiate members in terms of their

heterogeneity are:. rules for allocation and utilization of

resources;. methods and approaches utilized in

problem solving;. degree of adaptability shown in

negotiation and compromise; and. extent of inter- and intra-activity

interactions within the member

organization.

System analysis of a member deals with

evaluating the influence of these factors

primarily on its member organization

and through interactions on other

members of the supply chain. This is

accomplished by:

1 conducting a value analysis of primary

activities as depicted in Table I; and

2 identifying hierarchy of controls in the

decision-making process as depicted in

Table III, of the member enterprise.

Figure 6 illustrates the architecture of a

CSC member enterprise. The integration of

design components, that are described in

the section entitled, `̀ a distributed problem-

solving hypothesis for a cooperative supply-

chain system'', is evident in this diagram.

Decision-making models are aggregated

from the lowest (activity) to the highest

(member) component of the enterprise. The

transformation of material from one stage

to the next until final product is derived

occurs at the activity level. Transformation

in the order-life-cycle occurs at the business

level, as marketing and sales, order entry,

product design, production planning and

scheduling, manufacturing, and shipping

functions, respectively process the order.

Controls are passed at both inter

(`̀ between'' operations along the model

hierarchy), and intra (`̀ within'' operations

belonging to the same function) levels to

implement independent organizational

goals, policies, and objectives.

Cooperative supply-chain system:a group perspective

The organization of a group in a CSC

system is with the purpose of finding

homogeneity in the heterogeneous behavior

of supply-chain members (Malone, 1990).

This is accomplished by gaining mutual

commitment and converging on joint

intentions of members towards achieving

common supply-chain goals. Some norms

that enunciate this spirit of are:. allocation rules for sharing scarce

resources in the supply chain;. rules for cooperation and coordination;

Figure 5A supply-chain enterprise work design and methods improvement approach

[ 299 ]



. adoption of a problem-solving approach by

the group;. defining roles and responsibilities of each

member;. rules for negotiation and compromise; and. extent of inter- and intra-activity

interactions in the group.

System analysis of a group deals with

evaluating the influence of these factors on

the group enterprise. This is accomplished

by:

1 conducting a value analysis of primary

activities as depicted in Table II; and

2 identifying a hierarchy of controls in the

decision-making process as depicted in

Table IV, of the group enterprise.

Figure 7 illustrates the architecture of a

CSC group enterprise. The integration of

design components, that are described in

the section entitled, `̀ a distributed problem-

solving hypothesis for a cooperative supply-

chain system'', is evident in this diagram.

Decision-making models are aggregated

from the lowest (member business) to the

highest (group) component of the

enterprise. The transformation in the

order-life-cycle occurs at the member

business level as the order is processed, for

example, by marketing and sales function of

member enterprises, in the sequence they

add value to the product. Controls are

passed at both inter (`̀ between'' common

business operations of group and a member

along the model hierarchy), and intra

(`̀ within'' business operations belonging to

the same function) levels to implement

common supply-chain goals and objectives,

and policies.

Cooperative supply-chain systemmodeling

The main thrust of CSC modeling is based on

the principle that its architecture should be

domain independent, whereas its application

should be domain specific. Such a strategy

assures integration of disparate applications

to a common, yet generic architecture. Some

guidelines for implementing this strategy

are:. The structure of the conceptual model of a

CSC system must reflect inner workings

of its global (group) and local (member)

components.. The design of an application of CSC

should be based on a conceptual model of

the CSC architecture, but specific to

decision-making relevant to that

application.. Realization of objectives of a CSC

enterprise must be achieved through

implementation of a highly coordinated

set of strategies and policies at the global

and local levels. These should be

consistent with trends and directions

pursued by the industry for which the

supply chain is being designed.. The implementation of a CSC application

model must balance the issue of scope

versus focus.

The goodness of a CSC model for a problem is

judged by the ability of solutions to the

Figure 6A cooperative supply-chain member architecture

[ 300 ]



problem to satisfy necessary and sufficient

conditions posed by the problem. That is,. to satisfy necessary conditions, the CSC

model must reflect the requirements of

industry, as borne out of facts through

various methods of inquiry; and. to satisfy sufficient conditions, every

business strategy that facilitates

implementation of the CSC model, is a

candidate solution to the industry

supply-chain problem.

The above modeling concepts are further

elaborated by two representations of a CSC

system model:

1 A cooperative supply-chain system

decomposition model (DM) is depicted in

Figure 2 with notations in the Appendix,

section A (Taha, 1987). The technology

matrix Dj and corresponding resource

vector bj represent the independent

structure of the member. The technology

matrix Aj and the objective function

vector Cj denote the common structure of

the group derived from the homogeneity

of members. Controls embedded in the

technology matrix constitute

relationships between various strategies,

such as identified in Table III (marketing,

production planning, warehouse

operations, etc.), with regard to their goals

and objectives, and policies. These

controls are propagated as constraint

equations represented by the technology

matrix. A similar approach is applicable

for gathering group information,

per Table IV.

2 A cooperative supply-chain system dynamic

process flow model (DPFM), with notations

in the Appendix, section B (Hillier and

Lieberman, 1990). The supply-chain

network depicted in Figure 8 signifies the

structure of a member. It has source S

(supply) and sink N (demand) nodes with

transshipment L acting as an

intermediary node. Controls in this

network are implemented by modulating

(managing discrepancies) inputs (activity

flow rates) based on strategies identified

in Table III (marketing, production

planning, and warehouse operations) to

support goals, policies, and objectives.

Linking of member networks through

common control strategies (Table IV)

produces the structure for the group.

For both models, coefficients for various

decision variables in the technology matrix

are derived by the application of various

optimization techniques, such as methods

engineering and value engineering on

different operations of the enterprise across

the value-chain, depicted in Table I. A

similar approach is applicable for gathering

group information, per Table II.

Conclusions

In this paper, the problem of supply-chain

design has been approached as that of a

cooperative system design. The

composition of a cooperative system was

chosen as the collective behavior of its

constituents offers a unique architectural

framework for applying a distributed

problem-solving approach. The

decomposition process of the supply-chain

enterprise leading up to identification of

activities of its members reveals behavior

of member entities that is useful in

designing the supply-chain design problem.

Figure 7A cooperative supply-chain group architecture

[ 301 ]



A cooperative supply-chain system model is

presented incorporating the behavioral

traits of its members. These behaviors have

attributes that are essential to modeling

separation of members from their collective

identity, group, in a cooperative supply-

chain system. The unique feature of a

distributed problem-solving approach to

represent the behavior of a system through

its component entities offers opportunities

to model large-scale supply-chain systems

modularly.

ReferencesAberdeen Group (1996), Advanced Planning

Engine Technologies: Can Capital Generating

Technology Change the Face of

Manufacturing?, White Paper, February.

Bond, A.H. and Gasser, L. (Eds) (1988), Readings

in Distributed Artificial Intelligence, Morgan

Kaufmann, San Mateo, CA.

Chandra, C. (1997), `̀ A formal system analysis

methodology for a cooperative supply chain'',

Proceedings of the 2nd Annual International

Conference on Industrial Engineering

Applications and Practice, November 12-15,

San Diego, CA.

Durfee, E.H., Lesser, V.R. and Corkill, D.D. (1989),

`̀ Trends in cooperative distributed problem-

solving'', IEEE Transactions Knowledge Data

Engineering, Vol. 11 No. 1, pp. 63-83.

Gasser, L. (1991), `̀ Social conceptions of

knowledge and action: DAI foundations and

open systems semantics'', Artificial

Intelligence, Vol. 47, pp. 107-38.

Hillier and Lieberman (1990), Introduction to

Operations Research, McGraw-Hill,

New York, NY.

Lee, H.L. and Billington, C. (1993), `̀ Material

management in decentralized supply

chains'', Operations Research, Vol. 41 No. 5,

pp. 835-47.

Malone, T.W. (1990), `̀ Organizing information

processing systems: parallels between human

organizations and computer systems'', in

Zachary, W.W. and Robertson, S.P. (Eds),

Cognition, Computation and Cooperation,

Ablex, Norwood, NJ, pp. 56-83.

Moulin, B. and Chaib-Draa, B. (1996), `̀ An

overview of distributed artificial

intelligence'', in O'Hare, G.M.P. and

Jennings, N.R. (Eds), Foundations of

Distributed Artificial Intelligence, John Wiley

& Sons, New York, NY, pp. 3-55.

Nadler, G. (1970), Work Design: A Systems Concept,

Richard D. Irwin, Homewood, IL.

Niebel, B.W. (1993), Motion and Time Study,

Richard D. Irwin, Homewood, IL.

Poirier, C.C. (1999), Advanced Supply Chain

Management: How to Build a Sustained

Competitive Advantage, Barrett-Koehler

Publishers, Inc., San Francisco, CA.

Poirier, C.C., and Reiter, S.E. (1996), Supply Chain

Optimization, Barrett-Koehler Publishers,

Inc., San Francisco, CA.

Taha, H.A. (1987), Operations Research: An

Introduction, Macmillan, New York, NY.

Turbide, D. (1997), `̀ The new world of

procurement'', Midrange ERP, July-August,

pp. 12-16.

Tzafestas, S. and Kapsiotis, G. (1994),

`̀ Coordinated control of manufacturing/

supply chains using multi-level techniques'',

Computer Integrated Manufacturing Systems,

Vol. 7 No. 3, pp. 206-12.

Figure 8Illustration of a cooperative supply-chain dynamic process flow model

[ 302 ]



Appendix

A. Cooperative supply-chain systemdecomposition modelNotationAj ± technology matrix of the jth member

representing common constraints.

b0 ± resource vector for the jth member withrespect to common constraints.

Dj ± technology matrix of the jth memberrepresenting independent constraints.

bj ± resource vector for the jth member withrespect to independent constraints.

Cj ± vector of the objective functioncoefficients for the jth member.

Xj ± vector of decision variablescorresponding to jth member.

Problem formulation

Maximize, z = C1X1 + C2X2 + . . .. + CnXn

(Common objective)

subject to (constraints):

A1X1 + A2X2 + . . .. + AnXn = b0

(Common Constraints)

D1X1 = b1

D2X2 = b2

. . .

(Independent constraints)

DnXn = bn

Xj � 0, 8 j

(Non-negativity constraints)

B. A cooperative supply-chain systemdynamic process flow modelNotation

1 Variables

L ± level (stock) in the transshipment

node, denoted by a

R ± flow (rate) per unit of time t, denoted

by a

. RIN ± flow rate `̀ in'' through feedback

. ROUT ± flow rate `̀ out'' as feedforward

t ± time (period)

�t ± change at time t

xij ± average flow through arc i ! j

cij ± cost per unit flow through arc i ! j

uij ± arc capacity for arc i ! j

bi ± average net flow generated at node i

bi > 0, if node i is a source (supply node)

bi < 0, if node i is a sink (demand node)

bi = 0, if node i is a transshipment

intermediary node)

Auxiliary variables, denoted by a

GOAL ± desired level

CAP ± designed capacity

DISCg ± discrepancy with respect to

capacity

DISCc ± discrepancy with respect to a goal

DISCo ± discrepancy with respect to an

objective

OB ± desired objective

2 Problem statement:

Set up the flow model as a minimum cost

flow problem, that is,

MinimizeXn

i�1

Xn

j�1

Cij Xij

subject to (constraints):Xn

j�1

Xij ÿXn

j�1

Xji � b1; for each node i

and 0 � xij � uij , for each arc i ! j

3 Components of a system, represented as a

flow model:

Stocks (levels), rate of input/output,

source (S), sink (N), system boundary,

feedback/feedforward, goals, objectives,

policies

`̀ Intra'' flow

System boundary, capacity, feedback/

feedforward

`̀ Inter'' flow

System boundary, capacity, feedback/

feedforward

4 Conservation equations:

L (t + �t) = L(t) + (RIN ± ROUT) �t, [level

(stock) equation]

RIN (t) = F(L(t)), (rate equations)

and ROUT (t) = G(L(t))

where, F and G are some functions.

]

[ 303 ]



Documents

Enterprise architectural framework for supply‐chain integration