Content uploaded by Michael Glykas
Author content
All content in this area was uploaded by Michael Glykas on Dec 14, 2021
Content may be subject to copyright.
Knowledge and Process Management Volume 7 Number 1 pp 20–28 (2000)
"Case Study
A Case Study on Reengineering Manufacturing
Processes and Structures
George Valiris*
Department of Business Administration, University of the Aegean, Greece
Michalis Glykas
Engineering Department, Cambridge University, UK
Process thinking has expanded in recent years mainly due to the need for improved quality of operations. As
a result, there is a growing need for business modelling and analysis techniques that provide justification for
understanding and analysis of the organizational environment with the aim of developing appropriate
solutions for redesign. There is a lack of a systematic approach that can lead a process redesigner through a
series of steps for the achievement of process redesign. Most of the existing methodologies are based either
on real-life experience with little attention to the modelling and analysis of the business environment or vice
versa. In this paper we present a manufacturing company case study describing a specific framework for
business modelling and analysis. Copyright 2000 John Wiley & Sons, Ltd.
INTRODUCTION
Global competition is now the rule in most markets after the
end of protected domestic markets with guaranteed profits.
This has resulted in an exponential increase in the number of
new products and services as well as the speed at which
products and services invade the market. Competition has
also created a more demanding type of customer who is
constantly presented with a wide choice of good products
and services and who demands more from its suppliers than
in the past. For survival in today’s complex environment
enterprises have to present a high degree of flexibility and
timely response to surprises which could not be anticipated
in advance. They have to constantly re-invent themselves.
This demand for flexibility and timely response has led
many organizations to shift their focus from management
structures to business processes whose outputs are assessed
by their customers. Business Process Redesign (BPR) has
recently emerged in order to provide the necessary meth-
odologies that will allow companies to redesign their
business processes to successfully adapt to constantly
changing market demands. A plethora of BPR methodolo-
gies have been identified in the literature. However, a large
number of these methodologies were developed for other
purposes and were later relabelled to fall under the BPR
umbrella. Most of these relabelled methodologies appear to
have many limitations and there are only a few exemptions
where methodologies were developed solely for BPR
(Davenport, 1993;Harrington, 1991;Hammer, 1993;
Morris and Brandon, 1993;Petrozzo and Stepper, 1994;
Ould, 1992). However, even these methodologies are non
systematic (with the exception of (Morris and Brandon,
1993) and (Ould, 1992) attempts) and their emphasis is
more on hands-on experience and case studies.
A BPR methodology called Agent Relationship
Morphism Analysis (ARMA) has been developed (Glykas
et al., 1994a), aiming to combine accounting BPR principles
(efficiency, effectiveness, cost, etc.) with organizational–
theoretic concepts (roles, accountabilities, etc.) and some
powerful modelling techniques from Information System
(IS) development (Figure 1). The methodology is composed
of the following stages:
(1) Establishing the vision and objectives, the scope and
mode of BPR
*Correspondence to: George Valiris, Department of Business Admin-
istration, University of the Aegean, 8 Michalon Street, Chios 821 00,
Greece. E-mail: gval@aegean.gr
Copyright 2000 John Wiley & Sons, Ltd.
(2) Business modelling
(3) Business analysis
(4) Redesign
(5) Continuous improvement.
In the remainder of this paper we present a cigarette
manufacturing company case study describing ARMA’s
framework for business modelling and analysis.
THE CIGARETTE MANUFACTURING
COMPANY CASE STUDY
This research has been initiated by the most successful
cigarette manufacturing company in Greece, with 32.8% of
market share, to manage its risk management problems. The
company wanted to understand and analyse its business
processes with the aim of creating a more flexible organ-
izational structure and modernizing its operations in order
to survive in the domestic market and expand its business
internationally. The aim was the handling of economic risk
that companies face in today’s very competitive markets.
This need was not new for the company’s management.
However, in their previous attempts to redesign the organ-
ization with external support they were persuaded that the
risks involved in introducing process thinking in their
company were very high. Their main worry was the lack of
business modelling techniques that would provide the
necessary justification that the external party had properly
understood the organizational structure and processes, and
also the lack of business analysis techniques that are capable
of analysing the business models and developing solutions
for redesign at both the organizational and the employee
levels.
Consistent and simple modelling techniques were
required for overcoming the resistance of the company’s
employees who had argued in the past that redesign was
performed in an ad hoc and superficial way. The managers
wanted the employees to get involved and approve the
business models for two main reasons, first, for validating
(signing off) the business models of the existing situation
and second, for accepting more easily the proposals for
redesign in the future.
In the long run top management’s objective is to
expand their activities through extrovert but sound
policies in the same or related sectors to cigarette manu-
facturing as well as the sale of other products which use
the same distribution channels. In an attempt to achieve
the above long-term objectives top management is con-
centrating their efforts to improve the organization of the
company’s staff, modernize the technical facilities and
develop new parallel activities.
From the point of view of technology and products the
company is fully in a position to respond to market
demands. The major problem, however, is that is organiz-
ational structure lacks the flexibility required in order to
respond to competition and optimize productivity levels in
order to remain cost effective.
The need emerging from the business problems men-
tioned above is for business modelling and analysis tech-
niques that provide justification for understanding and
analysis of the organizational environment with the aim of
developing appropriate solutions for redesign.
Figure 1 The influence of existing methodologies in ARMA
CASE STUDYKnowledge and Process Management
Reengineering Manufacturing Processes and Structures 21
Our methodology was used for the modelling and
restructuring of five departments (domestic sales, export
sales, marketing, purchasing and information technology)
on an experimental basis. The cigarette manufacturing
company provided the ideal test-bed for BPR since it was
willing to (1) redesign its processes so that they are aligned
with its strategy, (2) introduce new IT systems, and (3) focus
on its customers. The primary objective of this case study
was to define the nature for the business opportunity for
improvement.
BUSINESS MODELLING IN ARMA
Business modelling in ARMA (Glykas et al., 1993c,1994b)
is achieved with the use of the three perspectives identified,
the structural, behavioural and process. The use of these
three perspectives provides insight into the relationship
between organizational structure and organizational pro-
cesses. Structural aspects are described in the structural
perspective whereas organizational dynamics are described
in the behaviour and process perspectives. In addition,
systems thinking, object-oriented and organizational theory
concepts are applied to the three perspectives. The connec-
tion between structure and processes in business modelling
in ARMA is shown in Figure 2. A technique called Agent
Relationship Modelling (ARM) has been developed for
modelling the structural perspective. The behavioural and
process perspectives are modelled in a technique called
Agent/Object Lifecycles (ALCs/OLCs). The notion of
ALCs/OLCs allows the modelling of the organization from
both the individualistic and the holistic views.
Most BPR methodologies lack the formal underpinning
to ensure the logical consistency of their business models.
The conceptual models developed in ARM and ALCs/OLCs
are given a formal underpinning in order to ensure their
logical consistency (Glykas et al., 1993a,d;Holden et al.,
1994).
The Agent Relationship Modelling Technique:
Modelling the Structural Perspective
The basic constructs of Agent Relationship Modelling
(ARM) are agents and objects. Agents represent the partici-
pants in contractual relationships and always include people.
Objects on the other hand represent either transactions
between agents or the resources that the agents utilize to
discharge their responsibilities.
A relationship is a general n-ary association between
agents and objects. Relationships, objects and agents can be
generalized to form classes which must have unique names.
One or both participating agent/object classes in a binary
Figure 2 The three perspectives in ARMA
Knowledge and Process ManagementCASE STUDY
22 G. Valiris and M. Glykas
relationship must have a responsibility. Cardinality con-
straints (min, max) have to be specified for both ends of a
relationship.
In ARM the agency concept is used in varying organ-
izational levels. In its general form it is a collection of
humans that constitute one of the parties in a contract. The
essence of a contractual relationship is that one or more
people employ one or more other persons as agents.
According to agency theory (Kaplan and Atkinson, 1991;
Bromwich, 1992) those who employ others are called
principals and those who work for others are called agents.
Following Johansson (1989) the whole organization, for
example, can be represented as one agent (macroagent). In
ARMA agencies can be decomposed into further agents and
resources. However, in their atomic form (not further
decomposable) an agent represents a person.
Relationships between agencies are called contractual.
ARM objects are either parts of the organizational agencies
or the subject of transactions between agencies during the
existence of the contract. They are subdivided into two
types: physical and logical. Physical objects are tangible
entities like computer terminals, printers, etc., whereas
logical objects include concepts like information, time, etc.
Objects are related to agents through functional and owner-
ship relationships. In functional relationships the agent cre-
ates, destroys, reads or modifies the state of an object while
performing an activity. An ownership relationship occurs
when a resource is utilized to carry out some of the agent’s
activities. In ARMA we believe that the existence of every
object should be linked to a specific contractual relationship
so that its existence is justified.
The functional, ownership and contractual relationships
are related to intentionality. Following Johansson’s (1989)
ideas, intentionality is a special form of connection between
subjects (agents) and objects or between subjects and
subjects. Functional and ownership relationships fall in the
former category whereas the contractual relationships
between agents fall in the latter.
In ARM objects and agents are represented by boxes and
relationships between them by rhomboids. Binary relation-
ships do not need to be given a name. Complex agents
(macroagents) are the result of the aggregation of some
other agents (which again may be complex) called partici-
pant agents. Similarly, complex objects result from the
aggregation of other objects called component objects.
Double-lined boxes represent complex objects and agents.
Double-lined thromboids represent complex relationships.
For example, Figure 3 depicts the top-level ARM diagram
for the cigarette manufacturing company as a complex
agency with its six participants that are in turn complex, the
new product development, the control and support, the
production, the preparation for production, the sales ac-
quisition and customer support. This is denoted by
the ‘has-part’,‘is-part-of’responsibilities undertaken by the
manufacturing company and the participant agents respect-
ively. The agencies with which the manufacturing company
interacts are of three kinds: the suppliers, the customers and
the authorities. Its participants are engaged with these
external agencies in a number of different contractual
relationships presented as complex relationships. The con-
tractual relationship ‘Consumable Goods Supply’, for
example, has two agencies as participants, the ‘Consumable
Goods Supplier’and the ‘Preparation for Production’. The
transaction that is exchanged in the contractual relationship
is the ‘Consumable Good’. The responsibilities of the
consumable good supplier in this contract are to ‘supply
consumable good’to the preparation of production agency
and to ‘find the consumable good’. The preparation for
production agency, on the other hand, is responsible to ‘pay
consumable good’to the supplier and to ‘transfer/store’the
consumable good. The cardinality constraint ‘1.N’in these
responsibilities denotes the fact that any number of such
payments and transfers can happen.
Following Checkland’s (1981) definition of emergence
some relationships between a complex agent/object and
other agents/objects can only be meaningful if the complex
agent/object is considered as a whole. The responsibility of
the complex agent/object is then known, as an emergent
responsibility (abbreviated to E). In Figure 3, for example,
the responsibility ‘calls for inspection in case of legal
problems’is meaningful only when the manufacturing
company is considered as a whole.
A complex agent/object might participate in some rela-
tionships only because one of its participants/components
participates in it. These responsibilities are called delegated
(abbreviated to D). An emergent responsibility obviously
cannot be delegated. The responsibilities of participants of
the manufacturing company in Figure 3 are all delegated to
their own participants. For example, the responsibility ‘pays
consumable good’of the preparation for production
agent is delegated to the production material purchasing
department agent that is one of its participants.
The Agent/Object Lifecycle (ALC/OLC) technique:
modelling the behavioural and process perspectives
Modelling behaviour in ARMA
A technique called Agent/Object Lifecycles (ALCs/OLCs)
has been developed for modelling the behaviour of organ-
izational agents and objects. An Object Lifecycle (OLC) can be
constructed for every object class defined in ARM. In Figure
4we present the OLC for the cigarette box. Every object in
the object class must be in a certain state represented by
circles. There are two special states in every OLC, the
never-exist and cease-to-exist. Every complete OLC should
start from the former and end in the latter. Every object can
be transformed from one state to another by an operation
(represented by round-cornered rectangles). An operation
can have more than one input and output states. If they are
the same a double-arrowed link is used (the operation
‘Special-order-preparation’in Figure 4). some operations can
operate on any state without modifying it, and these
CASE STUDYKnowledge and Process Management
Reengineering Manufacturing Processes and Structures 23
Figure 3 The ARM model of the manufacturing company
Knowledge and Process ManagementCASE STUDY
24 G. Valiris and M. Glykas
operations are modelled with a double-lined double-
arrowed line (the operation ‘inspect’in Figure 4). The
operation ‘New’is an operation that is performed by the
object class for the creation of the object instance repre-
sented by the OLC. Agent Lifecycles (ALCs) are represented
in the same way as OLCs but the concept of a state is
substituted by the concept of an agent’srole. The ALC
operations are the actions through which the agent
discharges its obligation to other agents (through access to
relevant resources) by maintaining or changing the state of
affairs. In Figure 5 we present the lifecycle for the driver
and in Figure 6 the lifecycle for the deliverer of the
manufacturing company’s products.
Modelling processes in ARMA
The process perspective in ARMA is concerned with the
identification and modelling of processes of importance in
the organization (business processes). The process perspec-
tive is based on functional composition of activities from
the lifecycles of the agencies and objects.
Functional composition in ARMA is achieved with the
use of operation schemata. We use the convention ‘Agency/
Object Name. Operation Name’to describe an operation in
the operation schemata. We assume that there must be an
aggregation relationship between component classes and an
aggregate class before operations of the aggregate class can
be composed by operations of the component classes. For
example, the operation of the product delivery agent in
Figure 3 is composed of activities from the cigarette box,
the driver and driver-assistant lifecycles in Figures 4–6,
respectively.
The ordering of operations in the operation schema may
or may not be specified. In the former case the ordering of
operations of component classes that are composed to
produce operations of their composite class is shown in a
top-down ascending order. For example, in the operation
schema describing the operation ‘delivery-to-customers’
first the product will be depalletized, then in some cases a
special order might be prepared etc. In cases of multiple
operations occurring at the same stage simple ‘and’/‘or’
constructs from logic theory are utilized.
BUSINESS ANALYSIS
The main focus of business analysis in ARMA is the analysis
of the business models with a view to developing solutions
for redesign. The emphasis is on why the operations and
processes at both the individualistic and holistic views of
the organization are performed the way they are performed.
The aim is to develop solutions that enhance the levels
of efficiency and effectiveness in these operations and
Figure 4 The object lifecycle for ‘Cigarette Box’
Figure 5 The agent lifecycle for the ‘Driver’of the manufacturing
company
Figure 6 The agent lifecycle for the ‘Deliverer’of the manufac-
turing company’s products
CASE STUDYKnowledge and Process Management
Reengineering Manufacturing Processes and Structures 25
Table 1 The contribution of ARMA’s business analysis techniques
Business
analysis
techniques
Alignment
between
business processes
and strategy
Relationship between
organizational
structure
and processes
Holistic and
individualistic
view of
the organization
Level of
communication,
coordination
and control
Market
metaphor
Level of
effectiveness
of activities
and processes
Level of
efficiency
of activities
and processes
Analysis
of the
conceptual
models
Analysis of the abstraction levels of the business models == =
Analysis of the dimensions of transactions == =
Principal–agent analysis == =
Management structure analysis == =
Mission/non-mission analysis =
Activity value added analysis ===
Fragmentation/concentration analysis ===
Equivalent salary analysis == =
Fractionalization analysis == =
Process cycle time analysis === =
Knowledge and Process ManagementCASE STUDY
26 G. Valiris and M. Glykas
processes. The dimensions of improvement can include
service, quality, timeliness, cost, etc.
A number of analysis techniques were found to be very
valuable for business analysis. Some of these modelling
tools were selected from the existing literature (Hammer,
1993;Davenport, 1993;Ip and Holden, 1992;Douma and
Schreuder, 1992;Fama and Jensen, 1993a,b;Williamson,
1975,1986), others were developed in order to satisfy
specific management needs. Different analysis techniques
are applied to the structural and behavioural/process
perspectives.
In existing BPR methodologies there exists a big division
in business analysis techniques due to the black-and-white
approach used in most cases. In some of them cost is the
central issue in others generic management or the successful
use of IT. As a result, business analysis techniques are
directed towards this central issue, ignoring any other
means of analysis.
In ARMA different analysis techniques from various
disciplines are applied in different perspectives (Table 1
presents ARMA’s business analysis techniques). These tech-
niques are influenced by organizational theories, IS devel-
opment and existing work in BPR giving business analysis a
more holistic approach. For example, principal/agent analy-
sis and the analysis of the dimensions of transactions are
influenced by agency theory and transaction costs econ-
omics respectively, mission/non-mission analysis and value-
added analysis by existing BPR work and abstraction level
analysis by IS development methodologies. The application
of some of the techniques like principal/agent and dimen-
sions of transactions are novel to BPR. Other techniques like
management structure analysis, concentration analysis,
equivalent salary analysis and fractionalization analysis are
performed from a different viewpoint. For example, man-
agement structure analysis is considered as unnecessary
since the focus of BPR is on business processes rather
than organizational structures. However, in ARMA this
technique is considered necessary in the structural perspec-
tive since it provides insight into the level of communi-
cation, co-ordination and control in the company before
redesign.
Business analysis in ARMA allows the process
redesigner to:
(1) Analyse the organization from both the holistic and
individualistic views
(2) Explore the relationship between organizational struc-
ture and processes
(3) Examine the alignment between business strategy, busi-
ness processes and employee objectives
(4) Examine the level of communication, co-ordination and
control in existing structure and processes
(5) Analyse the impact of creating a market metaphor for
organizational behaviour
(6) Analyse the level of effectiveness of activities and
processes
(7) Analyse the level of efficiency of activities and processes
(8) Analyse the conceptual models
In Table 1 we show how ARMA’s business analysis
techniques contribute to the achievement of the above
results.
CONCLUSIONS
The use of ARMA in the cigarette manufacturing company
case study was particularly successful. The driver in this case
study was the need for business modelling and analysis that
would provide understanding of the organizational process
and structure and develop justifiable solutions for redesign.
The modelling techniques of ARMA with their three
perspectives and their ability to model the organization
from both the holistic and individualistic views proved to be
well accepted by employees in practice. The main reason
was that the employees felt that their own structural and
behavioural requirements as well as their expected behav-
iour in the business processes they participated were
captured.
The business analysis techniques provided invaluable
solutions for redesign. The main benefit was that different
analysis techniques were applied in different perspectives.
For this reason, management and employees felt that
business analysis had taken into account both the needs of
individual employees and of the organization as a whole.
The cigarette manufacturing company wanted to incre-
mentally improve its existing processes. The scope and
mode of BPR provided a very good guide in the subsequent
stages and increased considerably management’s confidence
about the risks involved in introducing process redesign in
the organization. The business modelling techniques were
appreciated by employees and top management. The con-
tractual perspective in business modelling and redesign
where the employees were agreeing to the final models
proved to be very valuable.
BPR is still a new field of research that is in need of
applied research in organizations. The case study presented
here has contributed significantly in this direction and
provided normative/benchmark data for the future. Meth-
odologies with significant and specific normative data and
benchmark comparative information appear to have a key
competitive advantage in the marketplace. This information
provides powerful comparative points of reference, which
are typically of high interest and value to future case
studies.
REFERENCES
Bromwich, M. (1992) Financial Reporting, Information and Capital
Markets, Pitman Publishing, London.
Checkland, P.B. (1981) Systems Thinking, Systems Practice, Wiley,
Chichester.
CASE STUDYKnowledge and Process Management
Reengineering Manufacturing Processes and Structures 27
Davenport, T.H. (1993) Process Innovation: Reengineering Work
Through Information Technology. Harvard Business School Press,
Boston, MA.
Douma, S. and Schreuder, H. (1992) Economic Approaches to Organ-
isations, Prentice-Hall, Hemel Hempstead.
Fama, E.F. and Jensen, M.C. (1983a) Separation of ownership and
control. Journal of Law and Economics, Vol. 26, pp. 301–26.
Fama, E.F. and Jensen, M.C. (1983b) Agency problems and residual
claims. Journal of Law and Economics, Vol. 26/15, pp. 327–50.
Glykas (1993a).
Glykas, M., Wilhelmij, P. and Holden, T. (1993b) Verifiable object
oriented designs. In Technology of Object Oriented Languages and
Systems ’93 Conference Proceedings,1–6 August, Santa Barbara,
Prentice Hall, Englewood Cliffs, NJ.
Glykas, M., Wilhelmij, P. and Holden, T. (1993c) Object oriented
information systems development, In Proceedings of the 4th
Hellenic Conference, Patras, Greece, November.
Glykas, M., Wilhelmij, P. and Holden, T. (1993d) Formal methods
in object orientation. In Proceedings of the European Conference on
Object Oriented Programming (ECOOP) 93. Workshop on Appli-
cations of Object Oriented Formal Methods, Kaiserslautern,
Germany 26-30 July.
Glykas, M., Holden, T and Wilhelmij, P. (1994a) Modelling
organisational processes using the agent relationship morphism
methodology. In Proceedings of the 14th Hawaii International Con-
ference on Systems Science, Hawaii 1–4 January, IEEE Computer
Society Press, New York.
Glykas, M., Newton, A., Wilhelmij, P. and Holden, T. (1994b)
Information integration for intelligent petrochemical asset manage-
ment.IEEE Journal Transactions on Engineering Management.
Hammer, M. (1993) Reengineering The Corporation: A Manifesto for
Business Revolution. Nicholas Brealey Publishing, London.
Harrington, H.J. (1991) Business Process Improvement, McGraw-Hill,
New York.
Holden, T., Glykas, M., Wilhelmij, P. and Reynolds, B. (1994)
Lifetrack: organisational modelling for safety critical decision-
support. In F. Redmill and T. Anderson (eds), Technology and
Assessment of Safety Critical Systems, Springer-Verlag, London.
Ip, S. and Holden, T. (1992)A knowledge based technique for the
process modelling of information systems: the object lifecycle
diagram. In P. Loucopoulos (ed.), Proceedings of the 4th Conference
in Advanced Information Systems Engineering, Springer-Verlag,
Manchester.
Johansson, I. (1989) Ontological Investigations: An Inquiry into the
Categories of Nature, Man and Society, Routledge, London.
Kaplan, R.S. and Atkinson, A.A. (1991) Advanced Management
Accounting. Prentice-Hall, Englewood Cliffs, NJ.
Morris, D. and Brandon, J. (1993) Re-engineering Your Business,
McGraw-Hill, New York.
Ould, M.A. (1992) Process modelling with RADS. IOPener, Vol. 1,
No. 5, November.
Petrozzo, D.P. and Stepper, J.C. (1994) Successful Reengineering.Van
Nostrand, Reinhold, New York.
Williamson, O.E. (1975) Markets and Hierarchies: Analysis and Anti-
trust Implications, The Free Press, New York.
Williamson, O.E. (1986) Economic Organisation. Harvester Wheat-
sheaf, Hemel Hempstead.
Knowledge and Process ManagementCASE STUDY
28 G. Valiris and M. Glykas
