Info Systems J (2000) 10, 105–129 105

A conceptual framework for understanding

business processes and business process
modelling
Nuno Melão & Michael Pidd
Department of Management Science, The Management School, Lancaster University,
Lancaster LA1 4YX, UK, email: [Link]@[Link]

Abstract. It is increasingly common to describe organizations as sets of busi-

ness processes that can be analysed and improved by approaches such as busi-
ness process modelling. Successful business process modelling relies on an
adequate view of the nature of business processes, but there is a surprising diver-
gence of opinion about the nature of these processes. This paper proposes a
conceptual framework to organize different views of business processes under
four headings. It also aims at providing an integrated discussion of the different
streams of thought, their strengths and limitations, within business process mod-
elling. It argues that the multifaceted nature of business processes calls for plu-
ralistic and multidisciplinary modelling approaches.

Keywords: Business process management, business process modelling, business

process re-engineering, business processes, conceptual framework, modelling
approaches

INTRODUCTION

Recent years have seen an increasing interest in methodologies, techniques and tools to
support the (re)design of business processes under the banner of business process model-
ling (BPM). Many books, papers and theses have been produced, some of which are dis-
cussed in this paper. It is the subject of many undergraduate and postgraduate courses, as
well as the main topic of investigation in several research centres around the world. The hype
around BPM has also been fuelled by the enormous practitioner interest. For example,
Kettinger et al. (1997) surveyed 25 methodologies, 72 techniques and 102 tools. It is a require-
ment for many ISO 9000 quality programmes (Ould, 1995) and is at the crux of the imple-
mentation of many IT systems, such as workflow management systems (for example, see
Georgakopoulos et al., 1995) and enterprise resource planning (for example, see Robinson &

© 2000 Blackwell Science Ltd

 106 N Melão & M Pidd

Dilts, 1999). BPM includes a number of approaches, both hard and soft, which allow alter-
native process scenarios to be explored.
Although its origins can be traced back to the advent of scientific management around
the early years of this century, it was only with the introduction of business process re-
engineering (BPR) that BPM has emerged as a distinct field. Both BPR and BPM are based
on the notion that a business process is a fundamental element of analysis. However, there
seem to be almost as many definitions of business process as there are BPR books and BPM
techniques and tools. Most of the published literature provides rather superficial discussions
of this theme, although there are exceptions (e.g. Ould, 1995; Armistead & Rowland, 1996).
Given the prominence of BPR in recent years, such superficiality is a little puzzling. Further-
more, most of the literature on BPM and the nature of business processes is fragmented or
is restricted to a specific type of model and devoid of any thorough discussion of the field as
a whole.
The aim of this paper is to discuss the nature of business processes with a view to under-
standing how they can be modelled in order to improve their effectiveness. It groups different
notions of business processes under four headings. Although this does not provide a com-
plete classification scheme for business processes, it does provide a useful way of organiz-
ing and discussing different points of view about business processes and BPM. It also
suggests avenues for further research.

F R O M R E - E N G I N E E R I N G TO P R O C E S S M A N AG E M E N T

The early 1990s saw great interest in BPR. It was first described in North America by
Davenport & Short (1990) and Hammer (1990). It quickly became popular with management
consultancies (e.g. CSC/Index). Best-selling books appeared (e.g. Davenport, 1993; Hammer
& Champy, 1993), specialist conferences and journals (e.g. BPR Europe Conference, Busi-
ness Change & Re-engineering Journal) and network discussion groups (e.g. BPR Mailing
List BPR-L) were formed. Its rhetoric convinced many companies that dramatic breakthroughs
in performance could only be achieved by moving away from ‘functional hierarchy’ towards a
process-based paradigm using the power of IT. BPR became a panacea for many companies
around the world. The remainder of this section provides the setting for the rest of the paper
by discussing certain aspects of BPR that are the rationale for the emergence of BPM.

The paradoxes of BPR

Despite its popularity, BPR has many internal contradictions (Jones, 1994, 1995), which have
led to much confusion. From its early days, different management consultants used BPR as
a way to sell their own proprietary methods (Francis & MacIntosh, 1997; Grover & Malhotra,
1997). Inevitably, this led to confusion and disagreements. Responding to the claims of BPR
and the resulting confusion, the academic community criticized the powerful rhetoric and
vested interests of many consultancies for having no sound theoretical basis. Earl & Khan

© 2000 Blackwell Science Ltd, Information Systems Journal 10, 105–129

 Frameword for business processes and business process modelling 107

(1994) questioned whether there was much that was new in BPR. Mumford (1994, 1996) asked
similar questions, although focusing on some similarities between BPR and sociotechnical
design. Deakins & Makgill (1997) and Tinaikar et al. (1995) argued that the original
literature of BPR was essentially anecdotal, lacking serious and rigorous research to support
its recommendations and assertions.
The more recent literature strongly suggests that the radical approach to change, as
presented by the early BPR, is being softened by a number of lessons gleaned from both
successes and failures in the course of implementations. Perhaps, the best example is
Davenport & Stoddard (1994), who report on a field study and challenge some of the central
tenets of re-engineering in terms of ‘myths’. Their distinctive insight triggered others (e.g.
Burke & Peppard, 1995; Grover & Kettinger, 1995) and is the focus of current trends in the
field. The following summarizes these state-of-the-art issues around which debate has been
centred.

1. Novel vs. established. Hammer & Champy (1993) defined BPR as ‘a conceptually new busi-
ness model’ that has little or nothing in common with other improvement programmes.
However, various writers, for example Earl & Khan (1994), Jones (1995) and Peppard & Preece
(1995), have disputed this claim of novelty. Instead, they argue that the original
proposal for BPR linked together existing approaches in a novel way.
2. Radical vs. incremental. Hammer & Champy (1993) presented BPR as ‘the radical redesign
of business processes for dramatic improvement’. However, more recent empirical research
(e.g. Stoddard & Jarvenpaa, 1995) supports the view that, although the radical approach works
well in some companies, others, perhaps more conservative, prefer an incremental view, or
even something in between. For example, Cock & Hipkin (1997) compared the incremental
approach of TQM with the radical view espoused in early BPR. Even Hammer (1996), in
Beyond Re-engineering, recognizes his original mistake in asserting that the keyword of the
re-engineering concept was ‘radical’. This word, he admits, was responsible for the bandwagon
effect and excitement among managers. He now asserts that the most important word in the
definition is ‘process’.
3. Clean slate vs. existing process. Hammer & Champy (1993) stated that BPR has a ‘throw-
it-all-out-and-start-again’ flavour. However, Davenport & Stoddard (1994) and Stoddard &
Jarvenpaa (1995) assert that such clean slate change is rarely found.
4. Broad vs. narrow. The original notion of BPR was that it usually involved the redesign of
cross-functional business processes. However, research (e.g. Stoddard & Jarvenpaa, 1995;
Zairi & Sinclair, 1995) revealed that BPR initiatives within a single function can also provide
significant improvements and that this may sometimes be preferable as they are less demand-
ing and less risky.
5. IT-led vs. IT-enabled. Both Hammer (1990) and Davenport & Short (1990) stressed the
centrality of IT in BPR. Emerging research coupled with case evidence offers support for
another perspective – IT is an ‘enabler’ and a ‘creator of opportunities’. Coombs & Hull (1995),
Galliers & Baker (1995) and Guha et al. (1997) argued that BPR is possible without this central
support of IT.

© 2000 Blackwell Science Ltd, Information Systems Journal 10, 105–129

 108 N Melão & M Pidd

6. Mechanistic vs. holistic. The original BPR literature is heavily dominated by a hard sys-
tems approach and by machine metaphors of organization. But an increasing number of
authors call for a more holistic and softer approach in order to take into account strategic
and people issues. Earl et al. (1995) reported on field studies that demonstrated the danger
of a mechanistic approach. Galliers & Baker (1995) argued that soft OR, which assumes
that the world is problematic rather than given, provides one possible way of approaching
BPR.
7. Dramatic vs. modest. Hammer & Champy (1993) claimed that BPR involved ‘dramatic
improvements in critical, contemporary measures of performance’. Yet, empirical investigation
(e.g. Grover et al., 1995) reveals that BPR initiatives are delivering much less than they have
promised.
8. Top-down vs. bottom-up. Hammer & Champy (1993) insisted that BPR ‘never, ever happens
from bottom up’. Nevertheless, several studies found that the participation, commitment,
ownership and initiative from the front line were vital for many successful BPR programmes.
For instance, Willcocks & Smith (1995) explored some of the human dimensions of such
change.
9. Inspiration vs. methodology. Hammer (1990) and Hammer & Champy (1993) argued that
BPR ‘cannot be planned meticulously’. Instead, they believed that BPR depends largely on
imagination, creativity and experience. As with many movements, such inspiration may work
well for the pioneers; however, the next generation may need to see things formalized. More
recently, therefore, the field has seen the introduction of various methodologies and model-
ling tools to address BPR issues systematically. Kettinger et al. (1997) surveyed a number of
these methodologies, techniques and tools.

In view of these developments, it could be argued that the original concept of BPR is itself
being re-engineered to take a broader perspective. Davenport (1996), one of the pioneers of
BPR, claims that re-engineering is effectively finished, at least in the USA. He argues that, to
most business people in the US, re-engineering is associated with restructuring, with layoffs
and with failed change programmes. He admits that the reason for its failure is that people
do not like to be re-engineered. As discussed earlier, even Hammer (1996) has backed off
from the notion of radical change, instead focusing on the need to be process centred rather
than re-engineered.
Although the use of the term re-engineering may be dying, the focus on business processes
remains important. Business processes may be a natural way for work to be done in organi-
zations in order to create value for internal or external customers. This (for example, see Earl
& Khan, 1994; Ould, 1995) usually implies a distinction between different types of processes
based on value chain concepts, such as the following:

• core processes (which have external customers and include the primary activities of the
value chain);
• support processes (which have internal customers and concern secondary activities in the
value chain);
• management processes (which manage core and support processes).
© 2000 Blackwell Science Ltd, Information Systems Journal 10, 105–129
 Frameword for business processes and business process modelling 109

Figure 1. From re-engineering to process management and business process modelling.

It therefore seems reasonable to agree with Cypress (1994) that the ‘first generation’
re-engineering, which presented BPR as a new, radical, IT-led, mechanistic and inspira-
tional approach, is evolving to a ‘second generation’ process management. Process man-
agement views BPR as a hybrid, contingent, IT-enabled, holistic and systematic approach
(Figure 1). This evolution results from the recognition of an overemphasis on reductionist and
mechanistic aspects, in which crucial issues such as people and strategic issues were simply
ignored.
A process management perspective, as espoused latterly by BPR pioneers such as Hammer
(1996) and Davenport (1995), as well as by their critics (e.g. Kettinger & Grover, 1995; Peppard,
1996), is a continuum of approaches to process improvement. It includes radical (re-
engineering) and incremental (continuous improvement) perspectives, both of which should
be customized to the problem and context under consideration. The hallmark of process man-
agement is its focus on business processes. To support their improvement, the development
of process modelling approaches and tools is an important item in the new research agenda.

R E S E A RC H I N BU S I N E S S P R O C E S S M O D E L L I N G

Reports of research on BPM can be classified in various ways. Here, it is done under three
headings: reports by practitioners; attempts to develop theoretical positions; and discussions
of the nature of business processes. This illustrates that there are many ways of viewing the
activities and tools associated with BPM. In particular, it identifies the importance of suitable
and inclusive definitions of what might constitute a business process.

Reports by practitioners

The first group examines BPM as an activity by matching the typical process stages against
the tools and techniques commonly used by practitioners. Elzinga et al. (1995) derived a

© 2000 Blackwell Science Ltd, Information Systems Journal 10, 105–129

 110 N Melão & M Pidd

generic methodology by examining those used by different consultancies. From this, they
discussed a number of techniques and software tools that may help with implementing a par-
ticular stage or activity of that generic methodology. In like vein, Kettinger et al. (1997) sur-
veyed 25 methodologies, 72 techniques and 102 tools for supporting business process change
programmes. From this, they constructed a generic methodology, matching its stages with the
available techniques and tools. This leads to a conceptual framework for tool and technique
selection, based on the characteristics of the project. These studies confirm that practitioner
methodologies share many common features, despite the differences in philosophical orien-
tation towards IT, scale of change and people issues. Most consultancies have their own pro-
prietary methodologies and tools and stress a well-defined sequence of stages and activities.
A second group discusses the use of existing techniques for modelling business processes.
The appendix to Johansson et al. (1993) provided a brief overview of different process
mapping techniques with multifarious origins in work study, organization and method studies,
process control, process simulation, business modelling and systems engineering and analy-
sis. Similarly, Miers (1994) discussed and compared a number of techniques and their appli-
cation in modelling business processes, including flow charts, IDEF0, action workflow
diagrams and role activity diagrams. From an IS perspective, Darnton & Darnton (1997) sug-
gested a number of techniques and skills that should form part of the background of every
business process analyst.
A third group reviews and compares different software tools. Classe (1994), for example,
usefully reviewed 19 tools to support process improvement and redesign, including static,
dynamic, CASE and workflow. She illustrated how these tools are used through case studies
conducted in seven UK companies. She suggested that a contingent approach best describes
current practice, with the contingent adaptation based on a number of factors, such as the
goals of the project, scale and scope of change, the opportunity for IT support, culture, etc.
That is, although advanced modelling may sometimes be important, often simple word proces-
sors and spreadsheets are adequate for a BPM exercise. More formal BPM tools are used
not only to provide technical rigour and to test the impacts of alternative designs, but also to
support communication and participation. Similar, although more limited, studies were pre-
sented by Spurr et al. (1994), Bradley et al. (1995), Cory (1995), Cheung & Bal (1998) and by
Yu & Wright (1997). Most of these described the features and functionality offered by various
BPM tools and were quickly outdated after publication.
This examination of reports of practice suggests a lack of distinct methodologies, tech-
niques and tools to address the unique demands of modelling business processes. Instead,
practitioners have adapted techniques and tools from manufacturing, industrial engineering,
information systems, quality movement or human resource management and have applied
them to a BPM context.
Although this may sometimes be adequate, it is important to consider the nature of busi-
ness processes. Childe et al. (1995) reported a fieldwork exploration of BPR experiences, in
which they found that UK companies in general had an adequate understanding of the busi-
ness process concept and that further research should be directed towards other areas
instead. This paper is not so sanguine about this issue and argues that a better understand-

© 2000 Blackwell Science Ltd, Information Systems Journal 10, 105–129

 Frameword for business processes and business process modelling 111

ing of the nature and features of business processes is crucial if modelling is to be more
effective.

Theoretical views of BPM

Perhaps surprisingly, there are fewer academic papers on BPM than there are from those
proposing particular methodologies and approaches. An early theoretical view is that of Curtis
et al. (1992), who reviewed several modelling approaches and suggested a conceptual frame-
work for process modelling in the context of software engineering. This framework places busi-
ness processes in four positions: functional, representing the activities being performed with
relevant flows; behavioural, representing when and how the activities are performed; organi-
zational, representing where and by whom the activities are executed and what physical com-
munication mechanisms and storage are used; and informational, representing entities flowing
through the process, their structure and relationships.
Ould (1995), in proposing the role activity diagram (RAD) approach, also discussed mod-
elling issues and principles, although in the context of role activity diagrams. Scholz-Reiter &
Stickel (1996) and Scholz-Reiter et al. (1999) reviewed state-of-the-art issues on BPM from
an interdisciplinary point of view.
These apart, there are few significant attempts to develop theoretical positions on possible
approaches to BPM, possibly because the development of BPM has been driven by practi-
tioners rather than by academics.

Business processes – the core of BPM

Without suitable definitions of business processes, it is hard to develop suitable approaches,

whether theoretical or practical, to BPM. It would, however, seem that providing suitable
definitions is more difficult than might appear to be the case. Most of the literature simply
quotes (or adapts) the vague definitions put forward by the re-engineering pioneers. That is,
a business process is a set of related activities that are of value to a customer.
Moreover, most attempts to take this debate a step further have a rather mechanistic feel.
For example, Armistead & Rowland (1996) dedicated four chapters of their BPR book to busi-
ness processes, but their strong operations management bias, with its mechanistic empha-
sis, shone through. Similarly, Kock & McQueen (1996) reported an empirical study of 15
business processes in three companies but chose to stress structural features and informa-
tion flows, using ideas from industrial engineering and systems analysis and design. Both
studies argued that a business process is best viewed as a transformation of inputs from sup-
pliers into outputs to customers and that this transformation can be hierarchically decomposed
into subprocesses and activities. Although such views are not without value, they do rather
ignore the human side of business processes.
Moving away from such views, Ould (1995) argued that real-world processes are messier
than the input–transformation–output view might suggest and that the ‘neat and tidy’ hierar-
chy hid as much as it revealed. Instead, he argued that business processes are best viewed

© 2000 Blackwell Science Ltd, Information Systems Journal 10, 105–129

 112 N Melão & M Pidd

as networks in which a number of roles collaborate and interact to achieve a business goal.
This still has a less-than-human feel about it, but seems a step in the right direction. Going
further, Scherr (1993) emphasized people’s roles and relationships, but added a new dimen-
sion – a business process is seen as a set of closed loops of commitments.
Given this diversity of views, what are the implications for business processes? A useful
way to understand business processes is perhaps to regard them as multifaceted. Each per-
spective on business process is based on a set of assumptions about, for example, the nature
of organizational life, and these, in turn, affect the way in which approaches to BPM are devel-
oped. Each world view thus acts as a filter that allows us to see certain things but to miss
others.

F O U R P E R S P E C T I V E S O N BU S I N E S S P R O C E S S E S

Background

This section organizes views of business process around four themes, each with a different
emphasis and each illustrating important features of business processes. Two of these views
are similar to two metaphors found in Morgan (1997), whereas the other two are not really
metaphors, but viewpoints. Morgan’s metaphors have been widely used elsewhere and in
fields close to BPM. For instance, Peppard & Preece (1995) applied them to BPR, Pidd (1995)
to OR/MS and Walsham (1991) to IS. It seems obvious that views and understandings about
organizations are important when discussing BPM, as organizations provide the arena for
business processes. The four points of view suggested here view business processes as
deterministic machines, complex dynamic systems, interacting feedback loops and social
constructs.
The conceptual framework presented here does not attempt to match each author’s de-
finition of business processes with a single perspective. Even one person’s views can be mul-
tifaceted! Instead, it provides a useful way of organizing different points of views about
business processes and allows a discussion of the assumptions underlying BPM’s main
streams. Thus, a richer and wider picture is likely to occur. As with Morgan’s metaphors of
organization, each viewpoint sheds light on some elements, while it obscures others, and each
has strengths and limitations. However, when considered together, they produce a set of com-
plementary, yet competing, perspectives from which the nature of business process emerges.

Business processes as deterministic machines

The first view regards a business process as a fixed sequence of well-defined activities or
tasks performed by ‘human machines’ that convert inputs into outputs in order to accomplish
clear objectives (Figure 2). Not surprisingly, this standpoint is very close to Morgan’s bureau-
cratic machine metaphor, and it assumes that the nature of a business process is unques-
tioned, and its design is analogous to a technical engineering activity.

© 2000 Blackwell Science Ltd, Information Systems Journal 10, 105–129

 Frameword for business processes and business process modelling 113

Figure 2. Business processes as deterministic machines.

This view emphasizes the structure (tasks, activities and areas of responsibility), proce-
dures (constraints and rules of the work to be performed) and goals (nature of the output to
be obtained) of the business process being designed. The main criterion of good process
design is efficiency in the use of money, resources and time, subject to the constraint of sat-
isfying customers’ needs. Information technology plays an important role in this perspective:
automating, co-ordinating and supporting the re-engineered process. This accords well with
many structured processes found in stable manufacturing-type environments (e.g. order
fulfilment and fast food processes) and many bureaucratic paper-based processes found in
some service environments (e.g. credit application and back-office processes).
The notion that a business process is a deterministic machine can be traced back to Taylor’s
scientific management, in which manufacturing processes were made more efficient by an
analytical approach. This divided manufacturing processes into well-defined tasks to be per-
formed by interchangeable people. Each task was to be organized optimally by a manager
who would instruct and train the worker in the best way to do the task. This would lead to an
efficient overall manufacturing process.
In this vein, Davenport & Short (1990) defined a business process as ‘a set of logically related
tasks performed to achieve a defined business outcome’. Clearly, this notion, along with their
‘new industrial engineering’ metaphor, is symptomatic of a mechanistic view. Hammer &
Champy (1993) gave a similar definition, although highlighting customer orientation and cross-
functional activity. This is also the view of Armistead & Rowland (1996) and Kock & McQueen
(1996), whose focus was on the structural and operational features of business processes.
Early criticisms, which argued that BPR is the use of industrial engineering techniques applied
to office and service environments, were therefore inevitable (e.g. King, 1991).

Static business process modelling

As far as BPM is concerned, the view of a business process as a deterministic machine cor-
responds to the body of work underlying much of the hard and static approaches to BPM. In

© 2000 Blackwell Science Ltd, Information Systems Journal 10, 105–129

 114 N Melão & M Pidd

this, the stress is on mapping and documenting the flow of items, the activities, their logical
dependency and the resources needed. In recent times, many techniques from operations
management, operational research and information systems have been ‘repackaged’ to a BPM
context.
A frequently used technique, adapted from work study, is process flow charting and its exten-
sions (discussed by Johansson et al., 1993; Harrington et al., 1997). The simplicity of its basic
elements – activities and decisions – makes it easy to understand and communicate. Other
widely used techniques are IDEF0 and IDEF3, advocated for example by Hunt (1996) and
Mayer & DeWhite (1999). These are descendants of data flow diagrams with a functional,
structured approach. Their modelling constructs, inputs, activities, outputs, mechanisms and
controls in IDEF0 and units of behaviour, elaboration, referents, junctions, links, objects states
and state-transition arcs in IDEF3 reveal their mechanistic perspective.
Some approaches included in this perspective are more subtle. For example, rather than
focusing on procedures and data, Ould (1995) proposed a more process-oriented technique
called role activity diagrams (RADs). A RAD allows a business process to be modelled
diagrammatically through roles, goals, activities, interactions and business rules. Scherr (1993)
suggested another approach called customer–supplier protocol, based on speech acts
(Winograd & Flores, 1987). Here, a business process model is a network of a set of closed
loops of commitments between a customer and a performer involving four phases, namely
preparation, negotiation, performance and acceptance. Reijswoud et al. (1999) described
a similar approach. Unlike flow charts and IDEF techniques, both RAD and the customer–
supplier protocol are more appropriate for modelling business processes that involve the co-
operation of several entities and less appropriate for modelling complex routings. Examples
of other approaches include production engineering techniques (e.g. Doumeingts & Browne,
1997) and continuous improvement techniques (e.g. Schonberger & Knod, 1994).

Difficulties with a mechanistic view

Although these approaches have different strengths and limitations, it could be argued that
the mechanistic view has two major drawbacks. First, by assuming that business processes
can only be designed in rational and technical terms, it neglects human and organizational
issues. Empirical evidence (e.g. Willcocks & Smith, 1995) strongly suggests that IT-driven BPR
projects and a lack of attention to sociopolitical and organizational issues are major reasons
why so many BPR projects fail. This does not mean that a concern for technical and rational
issues is not important. Rather, it means that their consideration should not be overempha-
sized at the cost of the mismanagement of human and organizational issues. Writing on this
same theme, Morgan (1997) stated ‘by placing primary emphasis on the design of technical
“business systems” as the key to change, the majority of re-engineering programs mobilised
all kinds of social, cultural and political resistance that undermined their effectiveness’ (pp.
38–39).
The second criticism of this mechanistic view is that business processes are assumed to
be static. Static models are simplified representations of business processes at a particular

© 2000 Blackwell Science Ltd, Information Systems Journal 10, 105–129

 Frameword for business processes and business process modelling 115

point in time. As such, they ignore dynamic behaviour, which may change over time as a result
of resource competition, interactions or other sources of uncertainty. This does not mean that
static models are pointless. In fact, static models are very useful in understanding and repre-
senting the structural features of business processes and can be valuable means of com-
munication. However, business processes often display complex interactions that can only be
understood by unfolding behaviour through time.

Business processes as complex dynamic systems

Rather than viewing a business process as an assembly of interchangeable components, this

second viewpoint focuses on the complex, dynamic and interactive features of business
processes. The basic idea is close to Morgan’s metaphors of organism and flux and trans-
formation, in which an open system adapts to a changing environment in order to survive.
Whereas the mechanistic view focuses exclusively on structure and static objects, this view
emphasizes interaction and dynamic behaviour.
Viewed in these open systems terms, a business process can have inputs, transformation,
outputs and boundaries (Figure 3). A business process can then be defined as a set of sub-
systems: people, tasks, structure, technology, etc., which interact with each other (internal
relationships) and with their environment (external relationships) in order to fulfil some objec-
tive(s). Each subsystem can be seen as a system, which can in turn be hierarchically decom-
posed into further levels of detail. This in turn implies the definition of interfaces between
subsystems so that they are able to communicate with each other. The view of a business
process as a system, for example, is illustrated by Earl & Khan (1994), who said that the ‘inter-
dependent, interactive, boundary-crossing, super-ordinate goal conceptualisation of process
is essentially a systems view’ (p. 24).

Figure 3. Business processes as complex dynamic systems.

© 2000 Blackwell Science Ltd, Information Systems Journal 10, 105–129

 116 N Melão & M Pidd

The mechanistic perspective ignored important issues such as interactions with the exter-
nal environment, but this viewpoint pays much more attention to this. Therefore, effectiveness
(e.g. quality and service level) is likely to be a major design criterion rather than solely
efficiency. Another characteristic of this point of view is holism, stressing the behaviour of a
business process as a whole rather than its parts. For example, Hammer (1996) argued that
a sensible view of a business process ‘sees not individual tasks in isolation, but the entire
collection of tasks that contribute to a desired outcome’ (p. 11, authors’ emphasis). The use
of multiskilled and autonomous workers/teams to deal with a business process in a holistic
way illustrates particularly well how this holistic thinking can be put into practice. However, em-
pirical evidence, for example Zairi & Sinclair (1995), shows that it is not always possible to
approach business processes holistically, because it may be more risky and require more
resources than simply analysing a single component or set of components.

Discrete event business process simulation

Business processes are dynamic because of the interaction of their internal components and
because of the interaction of the process with its environment. Discrete event simulation (Pidd,
1998) provides a suitable way to model this dynamic behaviour in terms of entities (e.g. items
and resources) and discrete events (e.g. begin task and end task). The simulation model can
then be used to conduct ‘what if’ experiments, avoiding the need for building or disrupting the
real-world business process (Melão & Pidd, 1999). Modular, parsimonious construction of
models (Pidd, 1999) may usefully be employed by a modeller who needs to understand
complex business processes. Other, more complex approaches, such as high-level Petri nets
(e.g. Aalst & Hee, 1996), or less known approaches, such as qualitative simulation (e.g.
Nissen, 1996), could also be applied under this stream.
Vreede & Eijck (1998) described a discrete event simulation approach, Dynamic Modelling,
developed at Delft University. This is a problem-solving approach that uses simulation and ani-
mation techniques to analyse ‘as is’ business processes and evaluate future ‘to be’ business
processes. In particular, they see a business process as a network of interacting objects that
display dynamic behaviour. They argue that a business process can be modelled in terms of
three ‘parent’ object classes: items (composed of three ‘leaf’ object classes: message, product
and person); item processors (made of three ‘leaf’ object classes: actor, repository and link);
and actor actions (composed of two object ‘leaf’ classes: tasks and decisions). Each object
class has attributes and actions, which in turn are translated into blocks of SIMAN code. The
main strength of this approach is that the object–attribute–action world view represents a
natural way of modelling in the sense that it resembles real-world business processes.
However, the translation of the object-based conceptual modelling to the block-based SIMAN
model is not always straightforward. This problem has been addressed through the develop-
ment of an ARENA template (Eijck & Vreede, 1998).
Also in the context of BPM, Davies (1997) developed a data-driven simulator called SCOPE
that allows the simulation of information flows for high-volume back-offices. In an analogy with
manufacturing, he argued that office processes behave in a similar way to job shops, as

© 2000 Blackwell Science Ltd, Information Systems Journal 10, 105–129

 Frameword for business processes and business process modelling 117

documents pass through several functional departments for processing with routings that may
differ. However, he insists that office processes present unique modelling challenges, such as
high volume and mix of items. Furthermore, business processes have a mixed of front (core
processes) and back-office (support processes) components. Each item may require a dif-
ferent process, and there may be constant changes in workload and unpredictability intro-
duced by the human factor. SCOPE is capable of modelling the following elements: items,
teams, shifts, telephone calls, tasks, procedural flows and volumes. Although SCOPE is easy
to use, its main difficulty is that it is rather inflexible because the modeller is restricted to these
predefined modelling constructs and has no easy way of modifying or extending them. Like
the dynamic modelling approach, SCOPE seems to be more appropriate for modelling busi-
ness processes that display complex flows, but less appropriate for modelling business
processes that involve the co-operation of several entities.

Difficulties with a complex dynamic systems view

If the organic viewpoint is taken to an extreme, it also has weaknesses. First, it may lead to
the neglect of the sociopolitical dimension of a business process, as there is an implied belief
that a business process can only be approached in logical and rational terms. The human
aspect is only regarded as relevant as a resource for executing tasks, that is, the humanity of
the human is ignored. In this sense, this perspective is much like the previous one – the nature
of the business process and of its actors is taken for granted. Better process designs are, in
this view, based on an understanding of the logic of complex interactions with a view to
meeting the objectives set for the process in question.
The second problem is that such approaches obviously have a cost. The time and skills
required to build a dynamic computer model of simple systems may not add any value over
simple flowcharts or spreadsheets. Third, it ignores the feedback loops that may determine
the behaviour of many real-world business processes. Nevertheless, this viewpoint reminds
us that different subsystems of a business process interact to produce complex dynamic
behaviour.

Business processes as interacting feedback loops

This third perspective extends the organic viewpoint by highlighting the information feedback
structure of business processes. Like the organic perspective, the stress is on the complex,
interactive and dynamic features of business processes using systems thinking principles.
However, whereas the organic viewpoint focused on business processes with no intrinsic
control (i.e. open loop systems), this perspective claims that business processes are closed
loops with intrinsic control. This standpoint is thus an attempt to understand the dynamic
behaviour of a business process not in terms of individual components but in terms of inter-
actions between internal structure and policies.
The concept of a business process as a network of interacting feedback loops is shown
in Figure 4. This depicts a business process as flows (rates) of resources (physical or non-

© 2000 Blackwell Science Ltd, Information Systems Journal 10, 105–129

 118 N Melão & M Pidd

Figure 4. Business processes as interacting feedback loops.

physical) from outside its boundaries through a sequence of stocks (levels) representing accu-
mulations (e.g. materials) or transformations (e.g. raw material to finish product). The flows
are regulated by policies (decisions), which represent explicit statements of actions to be taken
in order to achieve a desired result (Pidd, 1996). These actions are taken based on informa-
tion, and this is where the notion of the information feedback loop comes into play. Such a
view of business process fits well with the notions of system dynamics discussed by Vennix
(1996) and van Ackere et al. (1993), who stressed the need to understand information feed-
back loops.

System dynamics

Within the BPM field, the view of a business process as interacting feedback loops is
supported by system dynamics modellers. Whereas a discrete event simulation is concerned
with modelling discrete state changes and individual entities, a system dynamics model of
a business process operates at a more aggregated level of abstraction, in which flow rates
are modelled as continuous variables. System dynamics models can be used in two ways
(Wolstenholme, 1990; Pidd, 1996; Vennix, 1996). First, in a qualitative mode, in which the
structural features of the process are made explicit through diagrams (e.g. causal loop and
flow diagrams). These in turn may become a basis for debate about the process behaviour.
For example, if a process were to be organized along particular lines, then such qualitative
models could provide insight into their potential stability. Second, system dynamics can be
applied in a quantitative mode by transforming the diagrams into a set of equations, so that
a simulation of the process can be conducted. This allows a modeller to provide quantitative
estimates of system effects.
Surprisingly, there are few reports of the application of system dynamics modelling in a
BPM context. The few research studies conducted in this area have been primarily concerned

© 2000 Blackwell Science Ltd, Information Systems Journal 10, 105–129

 Frameword for business processes and business process modelling 119

with justifying the use of systems dynamics in a BPM context and illustrating the potential
of its different uses, but going little further. This may suggest two things. The first is that the
BPM community may not be entirely convinced of the value of systems dynamics models.
The second is that this area remains relatively unexplored and thus presents rich opportuni-
ties for further research. For example, van Ackere et al. (1993) discussed the difficulties in
controlling the behaviour of logistical processes resulting from commonly occurring feedback
and delay structures using the classic ‘beer game’ to do so. They then showed the value of
continuous simulation in redesigning the structure of business processes in terms of decision
processes, physical processes and information channels. Fowler (1997) made a similar
argument.
Stevenson (1993) and Wolstenholme & Stevenson (1994) took a rather similar line, but
showed the value of both qualitative and quantitative system dynamics in mitigating the
misperceptions of feedback loop structure using the ITHINK system dynamics software. A
more critical line was followed by Davies (1996), who attempted to show the relevance of both
qualitative and quantitative system dynamics to the BPM field. He concluded by asserting that
‘challenges for the future development of business dynamics centre on ease of use, integra-
tion with other methods, and education’ (p. 241) in order to challenge some of the scepticism
about system dynamics. It might also be argued that the application of system dynamics to
BPM is limited by the difficulty of deriving generic implications of different internal structures
for the organizational design.

Difficulties with a feedback loop view

As with the previous perspectives, the view of a business process as interacting positive and
negative feedback loops has its limitations. First, when this perspective is taken to an extreme,
there is the risk of considering the human factor as only an instrument to be controlled or as
an instrument exercising control. System dynamics methods can be used in a sensitive, inter-
pretive mode, but the method itself carries no such guarantees. Indeed, it could be argued
that its roots in control theory and the apparent ease with which models may be constructed
could encourage an unthinking dehumanization of business process modelling.
Secondly, as Lane (1995) commented, ‘there is still too much belief and too little evidence’
(p. 617). Also taking a critical stance, Davies (1996) wrote, ‘system dynamics is easy to know
but impossible to apply’ (p. 241). That is, system dynamics approaches are easy to under-
stand at a superficial level, but may be difficult to use properly. Wolstenhome (1993) suggested
that system dynamics methods should be used in a qualitative mode as a means to elicit dif-
ferent views, to foster learning and to generate commitment. As far as BPM is concerned, this
may well be a sensible way to proceed, especially when time and difficulties in quantification
are critical constraints to the modelling exercise.
Perhaps these comments suggest that researchers might profitably devote some of
their efforts into engaged research (see Nandhakumar & Jones, 1995) in which they attempt
to use system dynamics methods in real-world business process modelling as action
research?

© 2000 Blackwell Science Ltd, Information Systems Journal 10, 105–129

 120 N Melão & M Pidd

Business processes as social constructs

Instead of seeing a business process as a predictable machine or as a dynamic organism

pursuing clear objectives, this fourth perspective emphasizes business processes as made
and enacted by people with different values, expectations and (possibly hidden) agendas. This
implies that business processes need not exist in the objective and concrete sense as in the
previous perspectives. Rather, they are abstractions, meanings and judgements that people
put on the real world, which result from a process of subjective construction of the minds of
people. The focus here is on subjective and human aspects of the business process.
From this standpoint, a business process can be defined in terms of different perceptions
constructed by various individuals and groups as a result of different frames of interpretation.
These frames, shaped by beliefs, values, expectations and previous experience, act as filters
enabling people to perceive some things but ignore others. For example, a production manager
may regard an order fulfilment process as a way to ensure that the orders are manufactured on
time, while a marketing manager may regard it as a way to satisfy a customer’s needs.
The existence of multiple (and often conflicting) views about what is going on and about how
the process is being and should be carried out means that a different view of change is
required. It implies that changes should result from a process of negotiation of conflicting inter-
ests, difficult though this process may be. The view of a business process as a social construct
fits well with strategic, less tangible processes, in which human activity is the major driver, such
as health, social and educational services. This view comes across quite strongly, for example
in Tinaikar et al. (1995), who called for a more humanistic social constructionist perspective in
an attempt to encourage the adoption of an alternative conceptualization of business process.

Soft business process modelling

It should be no surprise that this view of business processes as social constructs is closely
linked to a ‘soft’ strand of thinking about BPM. Unlike the previous viewpoints, soft models are
sense-making interpretive devices developed to generate debate and learning about how the
process is being and should be carried out. It should be noted that the technical view is not
entirely ignored. Indeed, techniques may be called in this perspective if the organizational
context requires it.
Several authors suggested the application of Checkland’s soft systems methodology (SSM)
to provide a more balanced approach to modelling business processes. For example, Galliers
(1994) observed that little attention has been given to exploring the role of soft modelling
in dealing with process issues and then goes on to outline an SSM-based approach to
undertake IS strategy/process change studies. Taking a more practitioner perspective, Patch-
ing (1995) showed how SSM provides a high-level, process-based language to approach
business process change from a holistic point of view. In the same vein, Chan & Choi
(1997) illustrated how SSM can be used to provide methodological support and an analytical
framework as well as to deal with ill-defined situations in a business process setting.
In these studies, SSM is used to represent a business process as a would-be purposeful
human activity system consisting of a set of logically interconnected activities through which

© 2000 Blackwell Science Ltd, Information Systems Journal 10, 105–129

 Frameword for business processes and business process modelling 121

Figure 5. Business processes as social constructs.

actors convert inputs into some outputs for customers. In addition, the business process
operates under certain environmental constraints, and monitoring and control is executed by
process owners. Moreover, the purposeful activity of the business process can be seen from
different angles (Figure 5). Soft BPM also relates to the humanistic view advocated by
Mumford (1994), in which a business process is seen as a set of interacting sociotechnical
systems. The objective is an integration of both social and technical needs through a partici-
pative approach. Unlike SSM, sociotechnical design places little stress on the cultural and
political environment.
The social constructionist viewpoint, when considered as the sole basis for modelling busi-
ness processes, may also have its own difficulties. First, the stress on ‘cultural feasibility’ may
impede the attainment of more efficient and radical designs. Secondly, this perspective alone
is unable to provide an objective, quantitative assessment of business process changes.
Finally, although it recognizes the political environment, it offers no way of dealing with it other
than the need to conduct several analyses of the type suggested by Checkland & Scholes
(1990). There have been a few reports, for example Lehaney et al. (1999), linking a con-
structionist view, using SSM, with hard modelling tools, but the benefits from doing so are not
yet clear in practice. Nevertheless, this does seem to be an avenue worth pursuing further.

D I S C U S S I O N A N D F U RT H E R R E S E A RC H

This paper stresses that BPM can be seen as a collection of methodologies, techniques and
tools supporting the analysis and improvement of business processes. It has been argued
that, to achieve greater modelling effectiveness in BPM, it is crucial to understand the nature
of business processes. For this purpose, this paper proposes that a business process may

© 2000 Blackwell Science Ltd, Information Systems Journal 10, 105–129

 122 N Melão & M Pidd

Figure 6. Business process views vs. paradigms.

be viewed from different and competing angles – deterministic machines, complex dynamic
systems, interacting feedback loops and social constructs.

Relating the perspectives with each other and with the paradigms

It is important to note, however, that these perspectives are not independent of each other
and that it is difficult to identify clearly where one perspective begins and the other ends. Figure
6 attempts to visualize these relationships and shows that these perspectives are best
regarded as inter-related facets of a multifaceted reality. The easiest linkage to grasp is,
perhaps, between the complex dynamic systems and the feedback loop perspective. Both
focus on the organic and dynamic features of business processes. However, the latter view
extends the former in an attempt to consider information feedback structure. Both perspec-
tives are related, in turn, to the mechanistic view, as they all tend to overlook social consid-
erations if taken to extremes.
On the other hand, the links with the social construction view are more difficult to perceive,
as it attempts explicitly to consider what other perspectives missed out – the human nature
of business processes. However, it is still possible to establish relationships with the complex
dynamic systems and the feedback loop perspective in the domain of SSM. All have in
common systems ideas, although they are used in different ways. Whereas in the complex
dynamic systems perspective, systems ideas are used to represent real-world business
processes, in the social construction view, systems thinking is applied as an intellectual device
to reason about people’s perspectives. In the feedback loop view, the use of systems ideas
is not so obvious, because it can be used in both a positivistic and an interpretive way.

© 2000 Blackwell Science Ltd, Information Systems Journal 10, 105–129

 Frameword for business processes and business process modelling 123

As mentioned earlier, each perspective has its own strengths, but also limitations when taken
to extremes. When combined, however, they provide, like metaphors, a range of complemen-
tary ideas from which one can better consider the nature of business processes. The combi-
nation of views proposed in this paper lead us to state that business processes have a mixed
and apparently conflicting nature. They have technical and social, tangible and intangible,
objective and subjective, quantitative and qualitative dimensions. Supporting this view, Grover
et al. (1995) reported on a large-scale survey of the implementation of BPR, in which they
found that business processes are best seen as dynamic sociotechnical systems. In the
broader field of IS, Checkland & Holwell (1998) used the well-established SSM-based action
research approach in an attempt to carry out a ‘conceptual cleansing’ of the field. They argued
that real organizations have a mixed nature as both social units, rationally pursuing well-
defined objectives, and (changing) social constructs, in which individuals or subgroups have
different interests and agendas. Or, as they neatly put it, ‘simultaneous would-be purposeful
“machines” and quasi-families’ (their stress) (p. 221).
Figure 6 also shows that each perspective makes different philosophical assumptions about
the nature of business processes and the relationship between the modeller and the business
process being modelled. For example, the roots of the mechanistic view in hard, technical
systems lead to an ontology in which the nature of the business process is objectively given,
is external and is composed of a number of discrete and tangible things. Epistemologically,
the role of the modeller is independently to abstract the real-world business process in order
to propose a cost-effective design that meets the objectives given. Similarly, in the complex
dynamic systems perspective, there is an implicit belief that business processes are ‘out there’
and consist of external, although interacting, entities. The role of the modeller is to under-
stand, external to the real-world business process, the complex set of interactions by mim-
icking its dynamic behaviour in order to propose a design alternative that meets the objec-
tives of effectiveness and efficiency.
Unlike the other perspectives, the social construction view regards the nature of business
processes as problematic, subjective and non-material. The modeller, more a facilitator than
a technical expert, cannot appreciate real-world business processes neutrally and needs to
work with the perceptions and meanings of the people involved in the process. The ontologi-
cal and epistemological positions of the feedback loop viewpoint are more difficult to discern
because of its hybrid nature. However, there may be a heavier inclination towards the posi-
tivistic stance as a result of its roots in control theory. Finally, it must be pointed out that the
philosophical stances suggested here should not be regarded as rigid. For example, there is
no reason why the dynamic complex systems perspective should not be applied in an inter-
pretive way, although this may be uncommon.

Implications for practice

What are the practical implications of these four perspectives on business processes and
BPM? One practical problem with contingent frameworks is that what seems beguilingly simple
on paper turns out to be rather difficult in practice, and a framework based on these four per-

© 2000 Blackwell Science Ltd, Information Systems Journal 10, 105–129

 124 N Melão & M Pidd

spectives is no exception. Nevertheless, it would be wrong to regard it as useless, just because

it may be hard, at this stage, to prescribe how it may be used by practitioners. There is a par-
allel with the approach to problem structuring in management science suggested by Pidd
(1996), using the ideas of Goffman (1974) and Schön (1982). Problem structuring is an
attempt to define the issues to be tackled in the modelling process. Pidd suggests that prob-
lems should be viewed as social constructs which, ‘like beauty, . . . are defined in the eyes of
the beholder’ (p. 72). Problem structuring, then, is a process in which different frames are
applied and various ‘problems’ are named, in order to provide a handle for their resolution.
Thus, problem structuring is a process of exploration in which multiple perspectives are useful,
and it may be that the same ideas apply in the early stages of business process investiga-
tion. It is unlikely that many management science practitioners will gladly dive into the detail
of Goffman’s sociology; nevertheless, it does provide useful insights that can be used to guide
thoughtful practice.
Perhaps the same is true of the four perspectives presented here? Few BPM and IS pro-
fessionals will bother to delve into the detailed assumptions that underlie the development of
the perspectives. Nevertheless, the multifaceted view does provide useful insights for prac-
tice. For example, if the BPM project focuses on the analysis and improvement of technical,
well-defined processes, then the techniques and tools underlying the mechanistic perspective
may well be appropriate. If the business process being analysed displays unpredictable,
complex interactions, then discrete event simulation techniques may be suitable. A business
process with a feedback loop structure would seem to call for systems dynamic approaches.
On the other hand, if the focus is on human, problematic processes, then the methodologies
under the social construction umbrella would seem to deliver a useful contribution.
To aid practitioners further in selecting the methodologies, tools and techniques appropri-
ate to the process under study, this paper also discussed the strengths and limitations of
different BPM approaches. In general, static approaches are useful for understanding and
communicating the structure of business process, but they lack a time dimension and, if
used in isolation, may ignore sociopolitical issues. Discrete event simulation approaches are
invaluable for understanding complex process interactions, yet they are resource consuming
and, used blindly, may neglect sociopolitical considerations. System dynamic approaches are
helpful for modelling business processes with feedback loop structure; however, they may
be rather mechanistic. Soft approaches are useful for addressing sociocultural issues, but
they lack the ability to provide an objective, quantitative assessment of business process
changes.
In addition, the results of this study stress the importance of pluralistic and multidisciplinary
modelling approaches. Willcocks & Smith (1995) suggested that many process improvement
programmes end in failure because the methodologies adopted are partial in their approach.
Commenting on the re-engineering fiasco, Davenport & Perez-Guardado (1999) argued that
process change programmes are better approached from different fronts in a multifaceted way
and illustrated this using an ecology metaphor. Hence, IS professionals can use these view-
points to construct more powerful modelling approaches. By thinking about alternative views
on business processes and using different BPM approaches, the modeller should be in a better

© 2000 Blackwell Science Ltd, Information Systems Journal 10, 105–129

 Frameword for business processes and business process modelling 125

position to capture the full richness and complexity of the situation, avoiding the limitations of
partial analysis.
Thus, any practical methodology should include provision for more than one viewpoint. For
example, the complex dynamic systems view could be strengthened by the social construc-
tion view, with simulation to provide a quantitative assessment of process changes and
with SSM to ensure social and cultural feasibility. Wastell et al. (1994), Wood et al. (1995),
Ackermann et al. (1999) and Warboys et al. (1999) proposed a number of approaches aiming to
link hard and soft modelling tools. Although their pertinence to a business process context needs
to be established, these examples do provide evidence of interest in such a methodology.

Implications for research

The findings of this study also have important implications for research. This paper contributes
to the BPM literature by proposing a conceptual framework aiming to classify alternative busi-
ness process views and to discuss different approaches to BPM. Of course, more viewpoints
could be considered, but these seemed to us the most relevant ones emerging from the lit-
erature at the time of writing. It extends the conceptual framework of Curtis et al. (1992) by
giving a place to soft modelling and by considering BPM from an holistic and multidisciplinary
perspective. It also relates to Checkland & Holwell’s (1998) investigation by providing the
theoretical underpinnings of the different schools of thought within BPM.
It is interesting to note the significant similarities between BPM and the discussion of IS by
Checkland & Holwell (1998). First, serious research in both fields is in an early stage, BPM
being younger than IS. Second, in both fields, theory lags behind practice. This is undoubt-
edly true for BPM, as the field has seen the introduction of many methodologies and tools
(often from consultancies), while serious research is only now emerging. Third, different
assumptions made about the nature of organizational life lead to different approaches in both
BPM and IS. Finally, the dichotomy between hard (the dominant) and soft (the emerging)
approaches seems to characterize most of the research in both fields.
The objective of this paper has been to derive a taxonomy of business process viewpoints
and to provide an account of the different streams of thought on BPM. It has argued that such
a framework will contribute to a clarification of the nature of business processes as well as
to an integrated discussion of a fragmented field. It has brought together a wide and diverse
literature, including OR/MS, IS/IT, software engineering, operations management, industrial
engineering and organizational research. Our conclusion is that business processes have a
multifaceted nature and, consequently, many BPM projects would be better off if they incor-
porated pluralistic and multidisciplinary approaches. An interesting avenue for further research
is to extend and develop fully the practical aspects of this framework.

AC K N O W L E D G E M E N T S

N. Melão is supported by the Portuguese Foundation for Science and Technology under
PRAXIS XXI.

© 2000 Blackwell Science Ltd, Information Systems Journal 10, 105–129

 126 N Melão & M Pidd

REFERENCES

Aalst, W. & Hee, K. (1996) Business process redesign: a Curtis, B., Marc, K. & Over, J. (1992) Process modeling.
Petri-net-based approach. Computers in Industry, 29, Communications of the ACM, 35 (9), 75–90.
15–26. Cypress, H. (1994) Re-engineering. OR/MS Today, 21,
van Ackere, A., Larsen, E. & Morecroft, J. (1993) Systems 18–29.
thinking and business process redesign: an application Darnton, G. & Darnton, M. (1997) Business Process
to the beer game. European Management Journal, 11, Analysis. International Thomson Business Press,
412–423. London.
Ackermann, F., Walls, L., Meer, R. & Borman, M. (1999) Davenport, T. (1993) Process Innovation. Harvard
Taking a strategic view of BPR to develop a multidisci- Business School Press, Cambridge, MA.
plinary framework. Journal of the Operational Research Davenport, T. (1995) Business process re-engineering:
Society, 50, 195–204. where it’s been, where it’s going. In: Business Process
Armistead, C. & Rowland, P. (eds) (1996) Managing Change: Re-Engineering Concepts, Methods and Tech-
Business Processes: BPR and Beyond. John Wiley, nologies. Grover, V. & Kettinger, W. (eds), pp. 1–13. Idea
Chichester. Publishing, Harrisburg.
Bradley, P., Browne, J., Jackson, S. & Jagdev, H. (1995) Davenport, T. (1996) Why re-engineering failed: the fad
Business process re-engineering (BPR) – a study of the that forgot people. Fast Company, 1, 70–74.
software tools currently available. Computers in Indus- Davenport, T. & Perez-Guardado, M. (1999) Process
try, 25, 309–330. ecology: a new metaphor for reengineering-oriented
Burke, G. & Peppard, J. (eds) (1995) Examining Business change. In: Business Process Engineering: Advancing
Process Re-Engineering: Current Perspectives and the State of the Art. Elzinga, J., Gulledge, T. & Chung-
Research Directions. Kogan Page, London. Yee, L. (eds), pp. 25–41. Kluwer Academic Publishers,
Chan, S. & Choi, C. (1997) A conceptual and analytical Massachusetts.
framework for business process re-engineering. Inter- Davenport, T. & Short, J. (1990) The new industrial engi-
national Journal of Production Economics, 50, 211– neering: information technology and business process
223. redesign. Sloan Management Review, 31 (4), 11–27.
Checkland, P. & Holwell, S. (1998) Information, Systems Davenport, T. & Stoddard, D. (1994) Re-engineering: busi-
and Information Systems: Making Sense of the Field. ness change of mythic proportions? MIS Quarterly, 18,
John Wiley, Chichester. 121–127.
Checkland, P. & Scholes, J. (1990) Soft Systems Method- Davies, M. (1996) Business dynamics: business process
ology in Action. John Wiley, Chichester. re-engineering and systems dynamics. In: Managing
Cheung, Y. & Bal, J. (1998) Process analysis techniques Business Processes: BPR and Beyond. Armistead, C. &
and tools for business improvements. Business Process Rowland, P. (eds), pp. 215–242. John Wiley, Chichester.
Management Journal, 4, 274–290. Davies, M.N. (1997) An Integrated Approach to Modelling
Childe, S., Maull, R. & Mills, B. (1995) UK Experiences Office Processes in Corporate Systems. Unpublished
in Business Process Re-engineering. The ESRC Busi- PhD Thesis, Department of Management Science,
ness Processes Resource Centre, Warwick University, Lancaster University, Lancaster.
Coventry. Deakins, E. & Makgill, H. (1997) What killed BPR? Some
Classe, A. (1994) Software Tools for Re-Engineering. evidence from the literature. Business Process Man-
Business Intelligence, London. agement Journal, 3, 81–107.
Cock, D. & Hipkin, I. (1997) TQM and BPR: beyond the Doumeingts, G. & Browne, J. (eds) (1997) Modelling
beyond myth. Journal of Management Studies, 34, Techniques for Business Process Re-Engineering and
659–675. Benchmarking. Kluwer Academic Publishers, Boston.
Coombs, R. & Hull, R. (1995) BPR as ‘IT-enabled organi- Earl, M. & Khan, B. (1994) How new is business process
sational change’: an assessment. New Technology, redesign? European Management Journal, 12, 20–
Work and Employment, 10, 121–131. 30.
Cory, T. (1995) Business Process Modelling Tool Products. Earl, M., Sampler, J. & Short, J. (1995) Strategies for
Sodan, Uxbridge. business process re-engineering: evidence from field

© 2000 Blackwell Science Ltd, Information Systems Journal 10, 105–129

 Frameword for business processes and business process modelling 127

studies. Journal of Management Information Systems, Hammer, M. (1996) Beyond Re-Engineering. Harper
12, 31–56. Collins, London.
Eijck, D. & Vreede, G. (1998) Simulation support for Hammer, M. & Champy, J. (1993) Re-engineering the
organisational coordination. In: Proceedings of the Corporation: A Manifesto for Business Revolution.
1998 Hawaiian Conference of Systems Sciences. Harper Business, New York.
Nunamaker, J. (ed.), pp. 633–642. IEE Computer Harrington, H., Esseling, E. & Nimwegen, H. (1997) Busi-
Society, Los Alamitos. ness Process Improvement Workbook. McGraw-Hill,
Elzinga, D., Horak, T., Chung, L. & Bruner, C. (1995) Busi- New York.
ness process management: survey and methodology. Hunt, V. (1996) Process Mapping: How to Re-Engineer
IEEE Transactions on Engineering Management, 42, Your Business Processes. John Wiley, New York.
119–128. Johansson, H., McHugh, P., Pendlebury, A. & Wheeler, W.
Fowler, A. (1997) The role of system dynamics in business (1993) Business Process Reengineering – Breakpoint
process analysis and management. In: Proceedings of Strategies for Market Dominance. John Wiley, New York.
the Business Process Track at the British Academy of Jones, M. (1994) Don’t emancipate, exaggerate: rhetoric,
Management Conference. Scarbrough, H. (ed.), pp. reality and reengineering. In: Transforming Organi-
157–175. The ESRC Business Processes Resource sations with Information Technology. Baskerville, R.,
Centre, Warwick University, Coventry. Smithson, S., Ngwenyama, O. & DeGross, J. (eds), pp.
Francis, A. & MacIntosh, R. (1997) The market, tech- 357–378. Elsevier Science, North-Holland, Amsterdam.
nological and industry contexts of business process Jones, M. (1995) The contradictions of business process
re-engineering in the UK. International Journal of Oper- re-engineering. In: Examining Business Process Re-
ations and Production Management, 17, 344–364. Engineering: Current Perspectives and Research Direc-
Galliers, R. (1994) Information systems, operational tions. Burke, G. & Peppard, J. (eds), pp. 43–59. Kogan
research and business re-engineering. International Page, London.
Transactions in Operational Research, 1, 159–167. Kettinger, W. & Grover, V. (1995) Toward a theory of
Galliers, R. & Baker, B. (1995) An approach to busi- business process change management. Journal of
ness process re-engineering: the contribution of socio- Management Information Systems, 12, 9–30.
technical and soft OR concepts. Infor, 33, 263–277. Kettinger, W., Teng, J. & Guha, S. (1997) Business process
Georgakopoulos, D., Hornick, M. & Sheth, A. (1995) An change: a study of methodologies, techniques, and
overview of workflow management: from process mod- tools. MIS Quarterly, 21, 55–80.
elling to workflow automation infrastructure. Distributed King, J. (1991) Rip it up. Computerworld, 15, 55–57.
and Parallel Databases, 3, 119–153. Kock, N. & McQueen, R. (1996) Product flow, breadth and
Goffman, E. (1974) Frame Analysis. Penguin Books, complexity of business processes. Business Process
Harmondsworth, Middx. Management Journal, 2 (2), 8–22.
Grover, V. & Kettinger, W. (eds) (1995) Business Process Lane, D. (1995) On the resurgence of management simu-
Change: Re-Engineering Concepts, Methods and Tech- lations and games. Journal of the Operational Research
nologies. Idea Publishing, Harrisburg. Society, 46, 604–625.
Grover, V. & Malhotra, M. (1997) Business process Lehaney, B., Clark, S. & Paul, R. (1999) A case of an inter-
re-engineering: a tutorial on the concept, evolution, vention in an outpatients department. Journal of the
method, technology and application. Journal of Opera- Operational Research Society, 50, 877–891.
tions Management, 15, 193–213. Mayer, R. & DeWhite, P. (1999) Delivering results: evolving
Grover, V., Jeong, S., Kettinger, W. & Teng, J. (1995) The BPR from art to science. In: Business Process Engi-
implementation of business process re-engineering. neering: Advancing the State of the Art. Elzinga, J.,
Journal of Management Information Systems, 12, Gulledge, T. & Chung-Yee, L. (eds), pp. 83–130. Kluwer
109–144. Academic Publishers, Massachusetts.
Guha, S., Grover, V., Kettinger, W. & Teng, J. (1997) Busi- Melão, N. & Pidd, M. (1999) Business process simulation:
ness process change and organizational performance: an overview. In: Proceedings of the UK Simulation
exploring an antecedent model. Journal of Management Society Conference. Al-Dabass, D. & Cheng, R. (eds),
Information Systems, 14, 119–154. pp. 126–132. UK Simulation Society, Cambridge.
Hammer, M. (1990) Re-engineering work: don’t automate, Miers, D. (1994) Use of tools and technology within a BPR
obliterate. Harvard Business Review, 90 (4), 104–112. initiative. In: Business Process Re-Engineering: Myth

© 2000 Blackwell Science Ltd, Information Systems Journal 10, 105–129

 128 N Melão & M Pidd

and Reality. Coulson-Thomas, C. (ed.), pp. 142–165. Schön, D. (1982) The Reflective Practitioner. How Profes-
Elsevier Science, North-Holland, Amsterdam. sionals Think in Action. Basic Books, New York.
Morgan, G. (1997) Images of Organization, 2nd edn. Schonberger, R. & Knod, E. (1994) Operations Manage-
Sage, Thousand Oaks. ment: Continuous Improvement. Irwin, Illinois.
Mumford, E. (1994) New treatments or old remedies: is Spurr, K., Layzell, P., Leslie, J. & Richards, N. (eds) (1994)
business process re-engineering really socio-technical Software Assistance for Business Re-Engineering. John
design. Journal of Strategic Information Systems, 3, Wiley, Chichester.
313–326. Stevenson, R. (1993) Strategic business process engi-
Mumford, E. (1996) Risky ideas in the risk society. Journal neering: a systems thinking approach using Ithink. In:
of Information Technology, 11, 321–331. Software Assistance for Business Re-Engineering.
Nandhakumar, J. & Jones, M. (1997) Too close for Spurr, K., Layzell, P., Jennison, L. & Richards, N. (eds),
comfort? Distance and engagement in interpretive infor- pp. 99–118. John Wiley, Chichester.
mation systems research. Information Systems Journal, Stoddard, D. & Jarvenpaa, S. (1995) Business process
7, 109–131. redesign: tactics for managing radical change. Journal
Nissen, M. (1996) Designing qualitative simulation of Management Information Systems, 12, 81–107.
systems for business. Simulation and Gaming, 27, Tinaikar, R., Hartman, A. & Nath, R. (1995) Rethinking
462–483. business process re-engineering: a social construc-
Ould, M. (1995) Business Processes: Modelling and tionist view. In: Examining Business Process Re-
Analysis for Re-Engineering and Improvement. John Engineering: Current Perspectives and Research Direc-
Wiley, Chichester. tions. Burke, G. & Peppard, J. (eds), pp. 107–116. Kogan
Patching, D. (1995) Business process re-engineering: don’t Page, London.
scare the horses. Management Services, April, 8–11. Vennix, J. (1996) Group Model Building: Facilitating Team
Peppard, J. (1996) Broadening visions of business Learning Using System Dynamics. John Wiley,
process re-engineering. Omega, 24, 255–270. Chichester.
Peppard, J. & Preece, I. (1995) The content, context, Vreede, G. & Eijck, D. (1998) Modelling and simulating
and process of business process re-engineering. In: organisational coordination. Simulation and Gaming, 29,
Examining Business Process Re-Engineering: Current 60–87.
Perspectives and Research Directions. Burke, G. & Walsham, G. (1991) Organisational metaphors and infor-
Peppard, J. (eds), pp. 157–185. Kogan Page, London. mation systems research. European Journal of Infor-
Pidd, M. (1995) Pictures from an exhibition: images of mation Systems, 1, 83–94.
OR/MS. European Journal of Operational Research, 81, Warboys, B., Kawalek, P., Robertson, I. & Greenwood, M.
479–488. (1999) Business Information Systems: a Process
Pidd, M. (1996) Tools for Thinking: Modelling in Manage- Approach. McGraw-Hill, London.
ment Science. John Wiley, Chichester. Wastell, D., White, P. & Kawalek, P. (1994) A methodology
Pidd, M. (1998) Computer Simulation in Management for business process redesign: experiences and issues.
Science, 4th edn. John Wiley, Chichester. Journal of Strategic Information Systems, 3, 23–
Pidd, M. (1999) Just modelling through: a rough guide to 40.
modelling. Interfaces, 29, 118–132. Willcocks, L. & Smith, G. (1995) IT-enabled business
Reijswoud, V., Mulder, H. & Dietz, J. (1999) Communica- process re-engineering: organizational and human
tive action-based business process and information resource dimensions. Journal of Strategic Information
systems modelling with DEMO. Information Systems Systems, 4, 279–301.
Journal, 9, 117–138. Winograd, T. & Flores, F. (1987) Understanding Com-
Robinson, A. & Dilts, D. (1999) OR & ERP: a match for the puters and Cognition. Addison-Wesley, Reading.
new millennium? OR/MS Today, 26, 17–21. Wolstenholme, E. (1990) System Enquiry. A System
Scherr, A. (1993) A new approach to business processes. Dynamics Approach. John Wiley, Chichester.
IBM Systems Journal, 32, 80–98. Wolstenhome, E. (1993) A case study in community care
Scholz-Reiter, B. & Stickel, E. (eds) (1996) Business using systems thinking. Journal of the Operational
Process Modelling. Springer-Verlag, Berlin. Research Society, 44, 925–934.
Scholz-Reiter, B., Stahlmann, H. & Nethe, A. (eds) (1999) Wolstenholme, E. & Stevenson, R. (1994) Systems think-
Process Modelling. Springer-Verlag, Berlin. ing and systems modelling – new perspectives on

© 2000 Blackwell Science Ltd, Information Systems Journal 10, 105–129

 Frameword for business processes and business process modelling 129

business strategy and process design. Management Biographies

Services, September, 22–25.
Wood, J., Vidgen, R., Wood-Harper, A. & Rose, J. (1995) Nuno Melão is a PhD student in the Department of
Business process redesign: radical change or reac- Management Science at Lancaster University. His re-
tionary tinkering? In: Examining Business Process search interests include business process modelling and
Re-Engineering: Current Perspectives and Research simulation.
Directions. Burke, G. & Peppard, J. (eds), pp. 245–261. Mike Pidd is Professor and Head of the Department of
Kogan Page, London. Management Science at Lancaster University. He is well-
Yu, B. & Wright, D. (1997) Software tools supporting busi- known for his work in computer simulation and is also the
ness process analysis and modelling. Business Process author of Tools for Thinking: Modelling in Management
Management Journal, 3, 133–150. Science, which discusses the links between hard and soft
Zairi, M. & Sinclair, D. (1995) Business process re- approaches in management science.
engineering and process management. Management
Decision, 33 (3), 3–16.

© 2000 Blackwell Science Ltd, Information Systems Journal 10, 105–129