TOWARDS AN INTEGRATED IS FRAMEWORK FOR THE

DESIGN AND MANAGEMENT OF LEAN SUPPLY CHAINS

Emmanuel Adamides, Nikos Karacapilidis, Hara Pylarinou, Dimitrios Koumanakos

Industrial Management and Information Systems Lab, MEAD, University of Patras, 26504 Rio Patras, Greece

Keywords: Web-based supply chain management.

Abstract: In this paper we present Co-LEAN, an integrated suite of software tools suitable for the design and

management of lean supply chains. In addition to providing full operational support in the planning and

execution of the lean supply chain, Co-LEAN supports internet-based collaboration in the innovation and

product design, manufacturing strategy, and supply-chain improvement tasks. The paper discusses the

information system support requirements of a lean supply chain, describes the main components and the

integration mechanisms of Co-LEAN and concludes with a brief description of its pilot use in a major

supermarket chain.

1 INTRODUCTION

The notion of lean supply chain has been a natural

extension of the concept of lean manufacturing.

Since mid-1990s, the philosophy and the toolset of

the lean approach were extended to cover the full

spectrum from the end customer to the raw material

suppliers (Hines and Rich, 1997; Rother and Shook,

1998; Hines et al., 2004). The holistic nature of lean

ideas calls for an equally holistic approach to the

supply, which deviates from limited in time, place

and scope activities, such as (optimised) supplier

selection, supply chain sourcing and supplier

development (Jones et al., 1997). Lean supply chain

management is contingent to the formation of

proactive relationships with all-tier suppliers, and

includes activities that extend beyond sourcing and

logistics. Consequently, in broad terms, lean supply

chain management can be defined as the

collaborative improvement of the entire product

lifecycle (from early design through planning,

production, maintenance and disposal) across both

functional and organisational boundaries. It involves

planning, executing and designing across multiple

partners in the supply chain to deliver products of

the right design, in the right quantity, at the right

place, at the right time.

In practical terms, a lean supply chain uses lean

manufacturing principles, such as 5S, visual factory,

takt time, pull, flow, etc. (Womack and Jones, 2003)

and Rate Based Planning and Execution (RBPE), a

way to smooth production and deliveries along the

supply chain (or network) through capacity planning

on the basis of the end-products bills of materials

(Reeve, 2002). Employing these lean techniques

along the chain, the objective is to reduce

redundancy in materials, information and knowledge

by providing them dynamically where they are

wanted, when they are wanted. To achieve this, a

number of challenges arise which include the

difficulty to see and manage cause-effect

relationships that occur across company boundaries

and are responsible for waste, the requirement to

break walls between normally insulated parties, and

the need to practice collaborative decision making

by developing a win/win philosophy of thinking

supported by agreements and trust between partners

(Sako, 1992, Cox, 2004; Hines et al., 2004; Evans

and Wolf, 2005). And, as the lines between

innovation/product development and manufacturing

and the supply chain are blurring, the lean supply

chain moves away from arm’s length contractual

relationships towards obligational contractual

relations (Sako, 1992; Kerrin, 2002) that both

support and are supported by collaboration in

product and process development, common vision

and shared goals, supply chain design, planning and

execution, as well as supply chain improvement

(keizen).

In practice, however, lean supply chain

implementations are rare, and collaboration in the

supply chain is still at an embryonic stage, mainly

concentrated on efforts for information exchange

among participants at the purely operational level, in

Adamides E., Karacapilidis N., Pylarinou H. and Koumanakos D. (2006).

TOWARDS AN INTEGRATED IS FRAMEWORK FOR THE DESIGN AND MANAGEMENT OF LEAN SUPPLY CHAINS.

In Proceedings of the Eighth International Conference on Enterprise Information Systems - SAIC, pages 35-42

DOI: 10.5220/0002459300350042

 SciTePress

the form of collaborative planning, forecasting and

replenishment (CPFR), vendor managed

replenishment (VMR) and synchronised supply

(Barratt, 2004; Holweg et al., 2005). This comes as a

consequence to the fact that organisations can

integrate their processes at the operational level with

relatively low difficulty, since this integration is

accomplished through codified information

exchanges. Other processes, such as innovation,

improvement and strategy, which are crucial to the

implementation of the lean supply chain, require the

capturing of tacit knowledge and the management of

complex interactions among the agents that hold it.

Collaboration in such issues requires either face-to-

face meetings, or the employment of advanced

information and communication technologies to

“virtualise” social interaction and support tacit

knowledge exchanges. In fast-changing industries

with world-wide supply chains, where challenges

and opportunities can arise at any instance, the

holding of meetings is quite difficult and may be

unproductive. Nevertheless, Internet can act as the

base enabling technology not only for information

exchange but also for more advanced collaboration

modes.

Towards this end, in this paper, we present an

integrated suite of software tools that can effectively

support the design and the management of a lean

supply chain. The suite consists of specially

developed tools to support collaboration among

supply chain partners in the innovation and product

development process, to assist in the development of

transparent and participative manufacturing strategy,

to enable the lean supply chain design and planning,

to facilitate its robust (adaptable) RBPE, as well as

supporting problem-solving and improvement

forums. Following, in Section 2, we review the ICT

requirements for an internet-enabled lean supply

chain. In Section 3 we present our suite of tools

explaining briefly each one’s functionalities. Finally,

before drawing our conclusions, Section 4 outlines

the operational environment of the suite’s pilot

application.

2 REQUIREMENTS FOR AN

INTERNET-ENABLED LEAN

SUPPLY CHAIN

Lean management concentrates on four areas which

are linked to different activities, at different levels.

The identification of value and the determination of

the appropriate structure for delivering it

(manufacturing strategy) are strategic

decisions/activities, whereas the specification and

the improvement of the value stream are activities

between the strategic level and the purely

operational one, to which the techniques such as

“flow” and “pull” belong. At the strategic level, the

identification and management of the value stream

starts with the definition of value from the customer

point of view. In the supply chain value is defined

not only by the end customers, but also by the

internal ones. Value has to do with products,

services and associated information, and is created

by innovative offerings that best meet the needs of

customers. This implies that, in addition to other

suppliers which fill organisational knowledge gaps,

customers should have an active role in the

innovation and product development process whose

final outcome is a collaborative effort.

Organisational knowledge gaps are the result of

the discrepancy between the knowledge an

organisation has and the knowledge it needs for the

solution of specific problems, including innovation

and product development. In filling these gaps, the

role of information technology is not only to

organise data into useful information, but also to

support the transformation of information into

organisational knowledge. Even more, since

innovation is a social process involving diverse

actors, there is a demanding necessity for

information and communications technologies to

support the knowledge flows among the relevant

actors and artefacts, in a way that enhances the

creation of new knowledge. As knowledge and

information flows are the key determinants of

successful innovation and new product development

processes (Tidd et al., 1997), their technological

support can augment their efficiency, effectiveness

and, consequently, their role as sources of

competitive advantage.

Once value has been defined in the form of

product (or service, or both), the means of producing

and delivering this value to the customer have to be

decided. This is the subject of operations/

manufacturing strategy which links the external

environment to the specific internal capabilities of

the firm(s) and its/their corporate strategy.

Manufacturing strategy is at a higher level from the

decision to implement a lean supply chain. Supply-

chain decisions belong to one of the four interrelated

decision areas of manufacturing/operations strategy

(capacity, supply chain/network, technology, and

development and organisation) (Slack and Lewis,

2002). The manufacturing strategy development is

an iterative process that involves participants from

different functions (marketing, product

development) and, sometimes, different

organisations. ICT can assist and leverage this

collaborative effort by not only helping strategists to

reach an agreed action plan effectively, but also to

ICEIS 2006 - SOFTWARE AGENTS AND INTERNET COMPUTING

augment learning the manufacturing strategy process

per se (Karacapilidis et al., 2006).

The definition of value and the alignment of the

operations with it act as inputs for the specification

of the value stream which extends along the whole

supply chain. Mapping, or better modelling and

simulating, the value stream facilitates the definition

of the supply chain and the identification of

initiatives for its improvement (reducing or

eliminating waste) (Rother and Shook, 1998;

Duggan, 2002). For these multiple

participants/stakeholders settings, collaborative

modelling and simulation technology (Miller et al.,

2001; Taylor, 2001) can enhance both the design and

redesign tasks by acting as the transitional object

(Morecroft, 2004) for improvement ideas, mind

frames and argumentations. Collaboration

technology can also be employed for general

problem solving and improvement along the supply

chain, thus fully supporting actions towards the

“continuous search for perfection” imperative of

lean.

Although the “flow” and “pull” imperatives are

associated with the design phase of supply chain (the

definition of structural and operational

characteristics of the system), they also entail purely

operational challenges which are concentrated at the

information layer and not with the knowledge one,

on which the value stream definition and perfection

components rest on. For supply chain planning and

execution, internet technology has been proposed as

a better alternative to attempts of legacy system

(ERP) integration along the supply chain (Kehoe and

Boughton, 2001). Since lean is contingent to stable

demand and operating conditions, at this level, it is

necessary to implement intelligent technologies to

assist in the efficient reconfiguration of the supply

chain when demand or internal operating conditions

change significantly. This can be done by exploiting,

automatically or semi-automatically, the

collaborative relations that exist among the partners

of the supply chain.

The lean supply chain characteristics can be

summarized in the requirements for synergy, as well

as knowledge and information integration across

activities in the entire product(s) life-cycle.

Information and communication technologies can

provide the basis for achieving these objectives.

3 THE CO-LEAN INTEGRATED

SUITE

The Co-LEAN suite consists of five interconnected

tools that operate on the internet for enabling the

implementation of the lean supply chain:

• Co-INOV supports the collaborative

innovation and product development

process (value specification),

• Co-MASS assists in the collaborative

development of transparent manufacturing

strategies at the individual node, as well as

at the entire network level (process of value

delivery),

• Co-SISC is used for collaborative

simulation-based supply chain/network

design (process/chain design and

improvement),

• Co-SOLVE supports problem solving and

improvement forums (process/chain

improvement) , and

• Co-LEAN-PE is the core of the suite and

provides facilities for rate-based scheduling

for individual products (value streams), as

well as for dynamic mixed-model

scheduling of product families by using the

relations that exist between product

component features and/or operations (flow

and pull operation).

Co-LEAN has been developed over the last three

years, initially as separate components that were

later integrated through Co-LEAN-PE. Co-INOV

and Co-SOLVE are based on Knowledge Breeder, a

web based software system that implements the G-

MoBSA (Group Model Building by Selection and

Argumentation) problem solving methodology

(Adamides and Karacapilidis, 2005a, Adamides and

Karacapilidis, 2005b). Knowledge Breeder is

collaboration supporting tool that uses a systemic

problem-knowledge representation scheme and an

evolutionary problem-resolution methodology that

supports the “breeding” (development) of the

knowledge necessary for the resolution of a specific

organizational issue. Co-SOLVE is a stand-alone

environment which is used as a platform for

discussion and argumentation forums centred around

specific lean supply chain improvement issues. On

the other hand, Co-INOV implements a more refined

and domain-specific version of Co-SOLVE. It can

support the innovation and product development

process in its entirety, enabling the gradual

“breeding” of innovation concepts towards their

realisation as products. In Co-INOV, the innovation

process is viewed as a sequence (not necessarily

being executed in linear fashion) of issue/problem

resolutions/solutions (Leonard and Sensiper, 2003),

in which cause-effect-like models are used to

represent organisational cognitive schemata of issues

and their possible resolution(s). The social and

knowledge dynamics of innovation are supported by

the co-evolution of models with respect to the shared

knowledge. A formal argumentation language is

used to facilitate participant interaction.

TOWARDS AN INTEGRATED IS FRAMEWORK FOR THE DESIGN AND MANAGEMENT OF LEAN SUPPLY

CHAINS

Figure 1: The functional structure of Co-INOV (based on

Knowledge Breeder)

The kernel of Co-INOV is its knowledge base that

stores models under consideration, as well as models

of already closed discussions. Any form of

electronic information (text, hypertext, drawings,

photographs, etc.), as well as direct links to

individual tools (e.g. to tools supporting the

engineering design phase of product development)

can be attached to the model(s) and transferred to

other actors (Fig. 1). Models are stored and selected

hierarchically using a meta-model of the issues

addressed (context definition). Users can upload the

current issues under consideration in which they

wish to be involved, see the current state of the

dialogue and contribute accordingly. They can then

move into other issues through the navigating meta-

model. By taking into account the structure of the

model, the arguments placed and the argumentation

rules, the inference engine of the computer-assisted

argumentation module, determines the defensibility

of each model of problem-solution. The interface of

Co-INOV (as well as that of Co-SOLVE) is in

hyper-textual form with menus associated to the

features provided, and diverse functionalities related

to visualisation (folding/unfolding of model

components, view/hide of inputs, creators, dates,

etc.)

Co-SISC is also a collaboration-enhancing

software environment that implements a language

and dialoguing structure specific to supply chain

design by means of simulation modeling

(Karacapilidis et al., 2004). It deploys a generic

systems-oriented language (activities, resources,

decisions and their parameters), thus enabling

different commercial simulation environments to be

easily used with. Again, argumentation on model

items and their attributes is supported by the system,

while storage and retrieval of

discussions/argumentations as well as simulation

models is possible. The construction, validation and

verification of the simulation model are

accomplished by a simulation specialist (technical

facilitator), whereas the other decision makers can

see its structure and results and experiment

accordingly with different parameters (Figures 2 and

3).

Figure 2: The operational architecture of Co-SISC.

The use of the Co-SISC environment in the lean

supply chain design/re-design process augments

shared understanding, transparency of individual

decisions and trust building along the supply chain

through an improved, in terms of quality and buy-in,

modelling and simulation process (Karacapilidis et

al., 2004). In connection with Co-SOLVE, it

provides the means for assessing activities, and

eliminating those which do not provide value to the

end customers (i.e. achieving leanness).

Figure 3: The main interface of Co-SISC.

Co-MASS is a computerised knowledge

management system for assisting in the formulation

Model

definition

Argumentation

and conflict

resolution

Innovation and product development tools

Idea

generation

and

processing

Process simulation,

parametric design

tools

Document

storage and

processing

Enterprise

systems and

external

linkages

Definition of

context

Model storage

and retrieval

Knowledge Breeder

ICEIS 2006 - SOFTWARE AGENTS AND INTERNET COMPUTING

of manufacturing and operations strategy

(Karacapilidis et al., 2006). It supports the social and

knowledge processes of collaborative manufacturing

strategy development by integrating a domain-

specific issue-modelling formalism based on the

resource based theory of the firm (Wernerfelt, 1984),

an associated structured dialogue scheme, an

argumentation enabling mechanism, and an efficient

algorithm for the evaluation of alternatives. As

sourcing is one of the principal decision areas of

manufacturing strategy, the direct provision of

knowledge from diverse sources across the supply

chain enhances the quality of the strategies produced

initially at the focal firm level, and then at the whole

of the supply network.

Finally, Co-LEAN-PE is the suite’s core module

that is responsible for the planning and execution

(rate-based) of the lean supply chain. It uses

forecasted and actual demand data, as well as

information from a continuously updated supply

database concerning products, suppliers and supplier

relations to produce a production and/or delivery

schedule (rates) for the focal company’s demand.

Demand information can be transmitted to all

suppliers in the network if needed. Schedules

(production and/or delivery rates) are calculated and

broadcasted to all nodes (Fig. 4, top window). The

feasibility of realisation of these schedules is then

examined by the software using the information

stored in the supply database. Additionally, suppliers

may contest schedules on the basis of their current

state. Should a schedule for a product is not feasible

by invoking the Relations Manager module, the

focal company may exploit the relations that exist

among products and suppliers at two levels: capacity

and capability. Capacity refers to the ability of a

different node that has the product in its current

schedule (or in its stocks) to supplement the initially

selected supplier with additional capacity, whereas

capability refers to the ability of producing the

product requiring additional capacity (the product is

not in the current production mix and there is no

stock). The additional cost burdens that the

exploitation of positive relations and the resolution

of negative ones imply is then calculated.

The Relations Manager uses conventional as well

as Artificial Intelligence techniques (representation

and coordination of plans) to represent and manage

the relations that exist between products and supplier

processes. It considers the products’ value streams

as plans to be executed with possible alternative

options for specific activities (e.g. two different

suppliers can provide the same part with two

different prices and delivery lead times), and by

using the relations that exist between supplier

operations and product characteristics tries to

coordinate the value streams so that optimal use of

resources along the whole of the supply chain is

achieved (von Martial, 1992). Nevertheless,

rescheduling and reconfiguration decisions

frequently require additional qualitative information

and discussion and argumentation using the features

of the other modules of Co-LEAN.

Figure 4: Co-LEAN-PE: the main screen (top window)

and lean supply chain construction (bottom window)

TOWARDS AN INTEGRATED IS FRAMEWORK FOR THE DESIGN AND MANAGEMENT OF LEAN SUPPLY

CHAINS

For the implementation of the proposed suite, we

have exploited a series of technologies supported by

the Microsoft’s .NET platform, such as ADO.NET,

XML Web Services, and Visual J#.NET

(http://www.microsoft.com/net/). The suite’s

architecture is shown in Fig. 5. As illustrated, access

to the tools of the Co-LEAN suite is provided

through a dedicated Web server. To use the range of

services provided, users only need a Web browser

(i.e., there is no need to download any specific

application at client side). Depending on the tool

used each time, users may exploit some built-in

templates and customize their working environment

according to their profile and collaboration

requirements. The interoperability of the suite’s tool

is achieved through the exchange of the appropriate

XML messages. There is also a dedicated SQL

server regulating the communication of the suite’s

tools with its proprietary database, model base and

knowledge base, as well as the communication with

remote databases, whenever there is a need to

retrieve data concerning particular “pieces” of the

supply chain. Connections with remote databases are

achieved through the available and well-tried

OLEDbControls of .NET platform.

4 USING THE Co-LEAN SUITE

The Co-LEAN suite is currently used, at a pilot

stage, in a major supermarket chain in Greece. The

objective has been to “lean” its fresh fruits and

Fig

ure

The architectu

re of the Co

LEAN suite

Supp

ly chain current state

Co-LEAN-PE

Schedules

(rates)

Co-LEAN-PE

Relations

Manager

Co-SOLVE

Co-INOV

Co-SISC

Co-MASS

Supplier

(remote)

databases

Collaboration

knowledge-

base

Forecasted

demand

Actual

demand

Suppliers

Figure 6: Coordination of Co-LEAN tools through the Relations Manager

ICEIS 2006 - SOFTWARE AGENTS AND INTERNET COMPUTING

vegetables supply chain, so that its image of

“freshness” in its products is enhanced. By using the

system described above, the company aims at

achieving small batch frequent deliveries in its shops

through three regional distribution centres. This

diverts from its previous operations where large

quantities of fruits and vegetables were delivered

and stored in one refrigerated warehouses before

being distributed to the selling points.

Using Co-LEAN, the supermarket chain can now

adjust deliveries to current demand, thus avoiding

costly and risky inventories at both the warehouse

and the shops. In addition, it can adjust the fruit and

vegetable picking rates of the suppliers. By feeding

Co-LEAN-PE with a weighted mix of forecasted and

actual demand, the required rates are calculated and

broadcasted to all suppliers, including the suppliers

of packaging materials (Fig. 4). Deliveries are

optimised using the Relations Manager to exploit

any slack delivery capacity of suppliers, thus

reducing the overall cost (Fig. 6). The Relations

Manager has also helped in overcoming the

inabilities of specific suppliers to deliver caused by

bad weather. Alternative suppliers have been

engaged in the most efficient way. The Co-SISC and

Co-INOV tools have been used in the initial design

of the supply chain, thus providing full transparency

to all parties involved. Co-SOLVE, Co-MASS and

Co-INOV have been used for developing in a

collaborative manner new packages which are

convenient for home deliveries of orders placed

though the internet. At next stage, the same

supermarket chain plans to use the Co-LEAN suite

in a more demanding area, that of its own-label

products produced by third-party manufacturers

under its control.

5 CONCLUSIONS

In this paper we presented Co-LEAN, an integrated

suite of software tools suitable for the design and

operation of lean supply chains. In addition to

providing full operational support in the planning

and execution of the lean supply chain, including

production and logistics optimisation, though its Co-

LEAN-PE tool, the suite supports internet-based

collaboration in the innovation and product design

(Co-INOV), manufacturing strategy (Co-MASS),

and supply-chain design and improvement tasks

(Co-SOLVE, Co-SISC). The pilot application of Co-

LEAN has provided an early indication of its

usefulness in a specific application area (retailing).

We are expecting that additional installations will

trigger modifications and adjustments that will, in

turn, lead to a more flexible and widely applicable

system.

REFERENCES

Adamides, E.D. and N. Karacapilidis (2005a). Knowledge

management and collaborative model building in the

strategy development process, Knowledge and Process

Management, 12(2), 77-88.

Adamides, E.D. and N. Karacapilidis (2005b). Information

technology support for the knowledge and social

processes of innovation management, Technovation,

Article in Press.

Barratt, M. (2004). Understanding the meaning of

collaboration in the supply chain, Supply Chain

Management, 9, 30-42.

Cox, A. (2004). The art of the possible: relationship

management in power regimes and supply chains,

Supply Chain Management, 9, 346-356.

Duggan, K.J. (2002). Creating Mixed Model Value

Streams: Practical Lean Techniques for Building to

Demand, Productivity Press, New York.

Evans, P. and B. Wolf (2005). Collaboration rules,

Harvard Business Review, 83(7), 96-104.

Hines, P. and N. Rich (1997). The seven value stream

mapping tools, International Journal of Operations

and Production Management, 17, 46-64.

Hines, P., M. Holweg and N. Rich (2004). Learning to

evolve: A review of contemporary lean thinking,

International Journal of Operations and Production

Management, 24, 994-1011.

Holweg, M., S. Disney, J. Holmstrom, and J. Smaros

(2005). Supply chain collaboration: Making sense of

the strategy continuum, European Management

Journal, 22, 170-181.

Jones, D.T., P. Hines and N. Rich (1997). Lean logistics,

International Journal of Physical Distribution and

Logistics Management, 27, 153-173.

Karacapilidis, N., E. Adamides and Ch. Evangelou (2003).

Leveraging organizational knowledge to formulate

manufacturing strategy, 11th European Conference on

Information Systems, Naples, June, CD of Conference

Proceedings.

Karacapilidis, N.I., Adamides, E.D. and Pappis, C. (2004).

An IS framework to support the collaborative design

of supply chains, Proceedings of KES 2004, 62-70

Kehoe, D. and N. Boughton (2001). Internet based supply

chain management: A classification of approaches to

manufacturing planning and control, International

Journal of Operations and Production Management,

21, 516-524.

Kerrin, M. (2002). Continuous improvement along the

supply chain: the impact of customer-supplier

relations, Integrated Manufacturing Systems, 13, 141-

149.

Leonard, D. and S. Sensiper (2003). The role of tacit

knowledge in group innovation, in: Choo, C.W. and

Bontis, N. (Eds.), The Strategic Management of

TOWARDS AN INTEGRATED IS FRAMEWORK FOR THE DESIGN AND MANAGEMENT OF LEAN SUPPLY

CHAINS

Intellectual Capital and Organizational Knowledge,

Oxford University Press, Oxford, 485-499.

Miller, J.A., P.A. Fishwick, S.J.E. Taylor, P. Benjamin

and B. Szymanski (2001). Research and commercial

opportunities in web-based simulation, Simulation

Practice and Theory, 9, 55-72.

Morecroft, J. (2004). Mental models and learning in

system dynamics practice. In: Systems Modelling:

Theory and Practice, Pidd M. (Ed), 101-126, Wiley,

Chichester, UK.

Reeve, J.M. (2002). The financial advantages of the lean

supply chain, Supply Chain Management Review,

March-April, 42-49.

Rother, M. and J. Shook (1998). Learning to See: Value

Stream Mapping to Add Value and Eliminate Muda,

The Lean Enterprise Institute, Brookline, MA.

Sako, M. (1992). Prices, Quality and Trust, Cambridge

University Press, Cambridge.

Slack, N. and M. Lewis (2002). Operations Strategy, FT

Prentice Hall, Harlow, UK.

Taylor, S.J.E. (2001). Netmeeting: A tool for collaborative

simulation modelling, International Journal of

Simulation: Systems, Science and Technology, 1, 59-

68.

Tidd, J., J. Bessant and K. Pavitt (1997). Managing

Innovation: Integrating Technological, Market and

Organisational Change, Wiley, Chichester, UK.

von Martial, F. (1992). Coordinating Plans of Autonomous

Agents, LNAI 610, Springer-Verlag, Berlin

Heidelberg.

Wernerfelt, B. (1984). A resource-based view of the firm,

Strategic Management Journal, 5, 171-180.

Womack, J.P. and D.T. Jones (2003). Lean Thinking:

Banish Waste and Create Wealth in your Corporation

(2nd edition, Revised and Updated), Free Press, New

York.

ICEIS 2006 - SOFTWARE AGENTS AND INTERNET COMPUTING