Basic Considerations on
Business Process Quality
Matthias Lohrmann1,2and Manfred Reichert1
1Institute of Databases and Information Systems, Ulm University, Germany
2KPMG AG Wirtschaftsprüfungsgesellschaft, Berlin, Germany
Abstract. Quality management practices in manufacturing and logis-
tics have led to proven results for organisations with respect to com-
petitiveness and profitability. At the same time, business process man-
agement not only addresses a comprehensive functional scope including
the ever more important administrative functions and business services,
but also contributes to realising potentials (e.g. in the field of process
automation) through extensive use of information technology. Integrat-
ing quality management with business process management concepts is
thus very promising from a business perspective. The evolution of a clear
understanding of business process quality constitutes a fundamental pre-
requisite for progress in this field as only a concise definition will point
out the issues that need to be addressed in more detailed research. This
paper aims at discussing basic considerations in this respect. We look
into requirements to be posed towards a definition of business process
quality, discuss various basic quality views and their fit with business
process management, provide a fundamental if not yet practically ap-
plicable business process quality definition, and examine related aspects
of business processes as well as related work. While this approach will
not lead to a final applicable definition of business process quality, its
contribution will lie in entering into a more systematic discussion to en-
compass the wide array of existing results that can be correlated with
business process quality and giving directions for future research.
1 Introduction
Since the 1980s, the impact of quality management (QM) practices, like design
for quality or statistical process control, on product quality and also on business
performance has been established through empirical studies [1,2,3,4]. Although
product quality and quality management practices are obviously by far not the
only determinant for business performance, most studies found these factors to
wield significant influence in terms of competitive advantage and, ultimately,
sustained profitability. It is therefore not surprising that QM has evolved into
one of the most pervasive management practices. An overview on its various
forms, which are often subsumed under the term Total Quality Management
(TQM), can be found in [5].
Traditional QM puts a strong emphasis on operations (for industrial businesses,
this means manufacturing and logistics) and partially services1as comprised in
the primary activities of an organisation according to Porter [8]. Beyond that
scope, business process management (BPM) in equal measure addresses other
primary and secondary activities that may be subsumed as administrative and
overhead processes, i.e. the materials and labour that do not directly enter into
end products. Since the relative weight of administrative and overhead processes
has been growing for decades in terms of the total cost base of organisations to
well beyond 50% in most industries [9,10,11], this characteristic has been a very
important factor to the success of BPM as a corporate practice. Moreover, BPM
provides organisations with a wide array of IT-supported methods and tools to
facilitate effective and efficient adoption.
Considering the proven merits of QM with respect to competitiveness as well
as profitability and the comprehensive functional scope addressed by BPM, we
expect that bringing these concepts together may lead to substantial benefits
for organisations by extending the field QM can be leveraged in. In addition,
it is promising to investigate how appropriate QM practices can be promoted
by integrating them with IT-based BPM methods. It will thus be rewarding to
assess how QM and BPM concepts can be aligned.
To be utilised effectively, QM always starts with specifying a notion of quality
the organisation aspires to achieve. Therefore, it is a crucial first step towards
QM for business processes and the objective of this paper to develop a notion
of business process quality which is based on QM insights, fits well with BPM
concepts and methods, and can be applied in a practical business environment.
In Section 2, we frame some practical requirements for a business process qual-
ity definition to ensure practical applicability of our results. Section 3 discusses
common definitions for the term business process as well as basic notions on
quality. In Section 4, we derive our fundamental definition of business process
quality and show that this approach leads to issues with respect to the require-
ments set out which we cannot easily resolve. On that basis, we enter a more
detailed discussion of quality-related aspects of business processes in Section 5.
This will facilitate to review existing work in relation to business process quality
in Section 6, and to identify remaining gaps and directions for future research
in Section 7.
1The quality of services provided to external customers is comparable to industrial
operations as well as “internal” administrative functions and constitutes a distinct
area of research. See, for instance, [6,7].
2
2 Practical Requirements Towards Definining Business
Process Quality
We judge practical applicability by matching our results with respect to defining
business process quality against three fundamental requirements. The require-
ments have emerged in our discussions with BPM practitioners from various
industries and will be subject to empirical assessement in the future course of
our research. We stipulate that these requirements constitute prerequisites for
QM to be effective in a BPM context:
1. Quality Measurability: According to Deming, not everything that is impor-
tant in management can be captured in visible figures [12, p. 121]. Neverthe-
less, a meta-analysis on empirical studies suggests that core QM practices,
which are mostly based on quantitative measures like statistical process con-
trol, positively influence quality as well as operational performance [13]. This
relationship has been recognised by many authors on quality management.
For instance, Crosby established quality measurement as one of his 14 steps
for quality improvement [14, p. 114]. Accordingly, management should strive
to base statements and conclusions regarding quality on concise quantita-
tive measures to facilitate appropriate controlling practices. We stipulate
that quality measures should be available on a ratio scale where possible
or at least on an interval scale. Ideally, quality measures are comparable
over three dimensions: time, processes and organisations. The dimensions
are based on the management practices of controlling over time, manage-
ment incentivation and benchmarking between organisations. Measurability
also subsumes that quality measures should not only be formally well-defined
but also applicable in practice; i.e. the effort involved may not exceed the
benefits to be gained (see for instance the IFRS framework, [15, paragraph
44]).
2. Quality Accountability: For each organisation, it is crucial to recognise that
the quality of its operations and products is determined by the actions of
its responsible personnel. In the end, quality is a people matter. Any notion
of business process quality should thus be devised in a way that clear orga-
nisational accountability for quality can be established. Quality assessment
results should be structured in a way to reflect accountable roles in an orga-
nisation, i.e. it should be possible to tell who is responsible for good or poor
quality without further analysis. This will be a prerequisite to utilise results
effectively for quality management and to drive actual improvements.
3. Quality Transparency: Statements and conclusions on quality should be trans-
parent and retraceable, i.e., accountable managers must understand the link
between status, actions and results. In a way, this requirement is an additio-
nal condition attached to the requirements of measurability and accountabi-
lity: Transparency can be reached if measure definitions and procedures are
not too complex or intricate. In addition, transparency is promoted if the
accountability link to organisational roles is straightforward, for instance via
corresponding job descriptions.
3
To illustrate the case for our three requirements, consider the following example
for a specific business process quality definition: assume that business process
quality in a service business is defined in a way that, at year end, one customer
is chosen at random and asked anonymously whether he perceived quality as
poor or good. This approach fulfils none of our requirements: results are not
measured on a ratio or interval scale, we do not know who interacted with the
customer and therefore is accountable for his or her quality perception, and it
is not transparent how the customer arrived at his judgement. Therefore, this
quality definition is not very useful for management purposes: For example, if
the quality next year is still “poor”, we do not know whether our efforts led to
any improvements or not, we do not know who of our staff we should approach
to discuss this result, and we do not know what exactly annoyed our customer.
To improve on this condition, we could interview one representative sample of
customers for each service team. We might ask the customers to state if they were
satisfied based on a clearly defined set of criteria. This would provide us with
a rate of satisfied customers per team, thus fulfilling all three requirements. We
could, for instance, track the rate of satisfied customers over time and identify
particularly good and bad teams. For each team lead, it would be transparent
how the customers arrived at their conclusions. However, as we will see later,
this practical applicability will still not be sufficient for an effective definition
of business process quality as other important aspects of business processes are
not considered!
3 Preliminary Considerations
In this section we shortly review common approaches to business processes and
aspects of quality. This background information is needed to obtain a basic
understanding on business process quality.
3.1 Common Business Process Definitions
The notion of business process first gained wide-spread recognition during the
early 1990s as the concept of business process reengineering became popular with
managers and consultants. Business process reengineering was advocated by au-
thors such as Davenport and Short or Hammer and Champy as a radical way
of changing and improving economically-oriented organisations [16,17,18,19]. In-
terestingly, the main motivation cited by the authors was American companies’
quest for a competitive response to the tremendous success of Japanese corpo-
rations at the time which has also been linked to their perceived lead in quality
management.
Davenport defines a business process as follows:
4
“A business process is simply a structured, measured set of activities
designed to produce a specified output for a particular customer or mar-
ket [...] A process is thus a specific ordering of work activities across
time and place, with a beginning, an end, and clearly defined inputs and
outputs.” [18, p. 5]
Concurrently, a quite similar view was developed by Hammer and Champy:
“We define a business process as a collection of activities that takes one
or more kinds of input and creates an output that is of value to the
customer.” [19]
While many authors have discussed these definitions [20], their core content has
basically not changed until today and has achieved wide-spread acceptance2. We
can therefore refer to the definition advocated by a well-established trade asso-
ciation. The Workflow Management Coalition (WfMC) defines the term business
process as follows:
“A set of one or more linked procedures or activities which collectively
realise a business objective or policy goal, normally within the context of
an organisational structure defining functional roles and relationships.”
[22, p. 10]
Overall, the authors agree that business processes consist of set of activities
which is linked to process input as included in the definitions above and to
process output or the realisation of a business objective. While process input
is not explicitly mentioned in the WfMC definition, each activity uses certain
resources of a kind, so the requirement for process input is implicitly comprised
in the notion of a set of activities.
In our further considerations, we moreover need to distinguish between a business
process model as an abstract notion and business process instances as concrete
enactments thereof. This distinction is supported by the WfMC definition of a
process instance as “the representation of a single enactment of a process” [22,
p. 16]. Business process models and business process instances are described in a
model proposed by Weske based on the UML class diagram notation [23, p. 94].
Hence, in our context we will interpret the term business process as the set of all
business process instances of the respective business process model, depending
on the context either in total or within a given timeframe.
We will also use two additional concepts: up- and downstream business processes
and artifacts. Up- and downstream business processes denote related business
2Contrary to the definition by Hammer and Champy and in agreement with Da-
venport, a certain ordering of activities or control flow today is generally assumed.
Notable exceptions are business process management approaches in the context of
case handling where additional degrees of freedom are required to cope with com-
plexity (see for instance [21]).
5
processes in process chains in the sense that an upstream process’s output consti-
tutes input for a downstream process. Note that aggregation and de-composition
of processes into super- and sub-processes is supported as well (see Section 5.1).
Artifacts denote all tangible objects or information items that are used by, cre-
ated in or altered by a business process.
3.2 Basic Notions on Quality
Since the 1950s, quality management has become one of the central management
concepts adopted by organisations globally. During that time, concepts and no-
tions for quality have evolved from the works of pioneers such as Shewhart,
Deming, Crosby, Feigenbaum, Juran and Ishikawa to standardised terminolo-
gies and methods that are propagated by trade and governmental bodies (for an
overview see [24]).
In terms of practical adoption, the notion of quality most widely spread today
has been developed by the International Organization for Standardization (ISO)
in the ISO 9000 series of standards [25]. As a set of norms in the area of quality
management for business applications, ISO 9000 has achieved broad acceptance
through endorsements by governmental bodies like the European Union and
the ISO 9000 certification scheme [26,27,28]. For a general notion of quality, we
can therefore resort to the definition given in the ISO 9000 series of standards:
quality denotes “the degree to which a set of inherent characteristics fulfils re-
quirements”.
This definition duly reflects a fundamental issue relevant for all approaches to-
wards quality management: determining quality is always based on a comparison
to an ideal, target or standard that sets requirements for the object in question.
In the following sections, we refer to this as quality standard. By defining a quality
standard considering the requirements set out in Section 2, we create a specific
definition of quality and thus enable quality management. Inversely, if we want
to apply quality management to a construct (e.g. a business process or an unit
of business process input), we have to define an appropriate quality standard
which is tantamount to developing a construct-specific quality definition.
In the course of the evolvement of quality management as a discipline, various
fundamental views on quality have been argued. These fundamental views ba-
sically correspond to differing classes of quality standards. As a basis to fur-
ther develop our understanding of quality with respect to aspects of business
processes, we discuss these views along a widely used classification by Garvin
[29,30]: quality can be discussed along “the transcendent approach of philoso-
phy”, “the product-based approach of economics”, “the user-based approach of
economics, marketing, and operations management”, and “the manufacturing-
and value-based approaches of operations management”.
Example 1. For an initial overview, Figure 1 illustrates differing views on quality
with a product quality example.
6
We also give an initial indication on the fit with our three requirements from
Section 2 and with BPM in general. Our appraisal regarding the requirements
set out focus on measurability and transparency because accountability is more
a question of how the quality management object has been defined rather than
how quality is interpreted.
3
User-based quality:
Polling customers showed their
appreciation of the fitness for
use
of
the
washing
machine
Product-based quality: 2
use
of
the
washing
machine
The machine achieves
clean laundry while using
little water and power 4
Manufacturing-based quality:
The washing machine
conforms to its
manufacturin
g
s
p
ecifications
g
p
Transcendent
quality
:
1
Value-based quality:
Matchin
g
the washin
g
Transcendent
quality
:
„It is clear that this washing
machine is well-made.“ 5
Washing Machine
g
g
machine‘s merits against its
price, buying it is a good deal
Figure 1. Quality views example
3.2.1 The Transcendent Approach
The transcendent view is the only approach that defines quality independently
of the perceptions or requirements of individuals or organisations, such as cus-
tomers, and therefore without an utilitarian or economic perspective. Instead,
according to this approach quality refers to “innate excellence” that is in princi-
ple applicable to all concepts in a BPM context and independent of any external
factors. While this view is deeply rooted in classical and modern Western philos-
ophy [31], it is consequently not in line with the ISO 9000 definition. Moreover,
BPM is aimed towards business process support [32]. Obviously, this implies an
economically motivated context. While we duly recognise its role in disruptive
innovation [33,34], we must conclude that transcendent quality is not appro-
priate for BPM applications, since BPM in itself is motivated by economic or
business-oriented targets. Moreover, the abstract nature of transcendent quality
or quality as excellence does not lend itself to practical quality assessment or
7
measurement [35] because results cannot be considered as objective, thus violat-
ing the measurability and transparency requirements. This is another obstacle
towards its practical applicability in a management context such as BPM.
3.2.2 The Product-based Approach
The product-based view centers on measuring concretely defined quantifyable
and desirable attributes in products [36]. In a BPM environment, we can substi-
tute the term product with the output of a business process, which also includes
end products. The higher the quantity of a quantifiable attribute that we find in
an unit of business process output, the better its quality is judged. Hence, this
approach can be directly linked to process output as a central construct in BPM,
and it supports management applications by enabling measurement. As an ex-
ample, consider the process of dejamming telephone landlines. A short cycle time
should hopefully be a desirable process output attribute for any provider and
will therefore constitute a valid quality measure in the product-based quality
view.
However, the product-based approach in a BPM context addresses process out-
put only. In a BPM context, this is not fully satisfactory as process input, process
execution and therefore the economic viability of business processes are not con-
sidered at all which is clearly not sufficient in a business context. Product-based
quality management always needs to be suitably complemented. An additional
issue relates to the relative weight of attributes considered for quality assessment
if more than one product characteristic is analysed. For instance, low weight and
high stability are attributes that are both valued in bicycle frames. How we judge
the relative merits of a heavy but robust frame against a light but fragile one
depends on the relative weight we allot to both characteristics, which may con-
stitute a major issue if our customers judge differently.
Thus, we can conclude that measurability according to this approach is typically
given because only well-defined quantifyable attributes need to be considered. On
the other hand, in many cases transparency is impeded: the relative weighing
of multiple attributes can usually not be done analytically, but needs to be
estimated.
3.2.3 The User-based Approach
The user-based view takes the product-based approach one step further by re-
placing measurable attributes of products (or process output) with the satisfac-
tion of a user or customer.
In principle, the same conclusions apply as to the product-based view regard-
ing measurability and applicability in a business context. The major difference
is that this approach is more inclusive and leads to more effective quality as-
sertions: not only selected, but all product or output attributes are considered,
8
and criteria are weighed to reflect user satisfaction; i.e. there is only one scale of
measurement. This allows for analytically well-founded normative statements on
quality. In the product-based approach, this is only possible if just one product
attribute determines quality. On the other hand, the weighing of product or out-
put attributes to effectively reflect user preferences (e.g. across a broad customer
base) constitutes an additional layer of complexity that is not easily resolved.
This as well as the alternative approach of polling users directly severely impedes
traceability.
3.2.4 The Manufacturing-based Approach
The manufacturing-based view focuses on conformance to specifications instead
of achieving optimum measures for certain product attributes or user satisfac-
tion. Similar to the product- and user-based approaches it is output-centered,
but implicitly recognises differences between optimum user satisfaction and as-
pired attribute values. It can be argued that the ISO quality definition has been
derived from this view on quality as it also stresses conformance to requirements.
The rationale behind this is that optimum user satisfaction or optimum at-
tribute values might not be economically sensible from the organisation’s point
of view. Insofar, the approach indirectly incorporates economic considerations
which corresponds well with the context of BPM. Moreover, approaches towards
manufacturing-based quality mostly include engineering (“design for quality”)
and production control, thus managing quality not only on the basis of process
output. With respect to measurability and transparency, the manufacturing-
based approach features the same advantages regarding measurability as the
product-based approach, but also the same issues regarding comprehensive qual-
ity measurement across multiple attributes and transparency.
Quality in the manufacturing-based approach also lacks a direct link to com-
mon quality expectations: under the manufacturing-based approach, it is easily
conceivable to have high-quality process output that does in no way satisfy the
expectations of customers or users, but fully conforms to specifications. From a
business perspective, high-quality manufacturing is therefore not sufficient but
needs to be complemented by defining high-quality specifications to conform to.
3.2.5 The Value-based Approach
The value-based view on quality incorporates the economic environment of or-
ganisations even more as it defines quality not only in terms of product attributes
and user expectations, but puts these in relation to the cost or price involved.
According to this view, good quality is achieved if a product or process output
does not only meet expectations, but also comes at a reasonable cost [37, p. 1].
This reflects the inclusive and economic nature of BPM best as it considers re-
sources consumption as well as output, and additionally incorporates economic
considerations.
9
However, Garvin rightly notes that this view on quality is hard to apply in
practice: it is difficult to comprehensively evaluate process input and output in
economic terms, which leads to issues with our measurability and transparency
requirements. This can be illustrated with the following examples:
–If a process delivers output that is not directly sold in the marketplace, its
value to the organisation cannot be trivially measured. An example for this
are research and development processes that do not lead to new products.
–Externally procured process input or upstream process output as input for
downstream processes cannot be valued at cost without further considera-
tions. For instance, supplier selection might not have been fully effective, or
upstream processes might not be designed optimally.
–Risks associated with process input or process execution are difficult to ap-
praise. For instance, a supplier might provide process input at a better price
but with added risk of not being able to deliver in time.
3.2.6 Summary: Views on Quality and BPM Requirements
Based on Garvin’s structure of quality approaches, we can summarise initial
considerations with respect to quality views and our requirements from Section
2 as well as the general BPM context in Table 1.
Quality View Fit with BPM Context Fit with Measurability and
Transparency Requirements
Transcendent
approach
No fit: does not correspond to a
business context
Not measurable
Not transparent
Product-based
approach
Limited consideration of economic
aspects
Measurability given
Limited transparency in case of
multiple attributes
User-based
approach
Limited consideration of economic
aspects
Measurability given
Severely limited transparency in
case of multiple attributes
Manufacturing-
based
approach
Indirect consideration of economic
aspects by definition of
economically viable specifications
Measurability given
Limited transparency in case of
multiple attributes
Value-based
approach
Consideration of economic aspects
by reflecting cost / value relation
Limited measurability due to
complexity
Limited transparency due to
complexity
Table 1. Quality views vs. BPM context and quality assessment requirements
10
4 Fundamental Business Process Quality
According to Section 3.2, we can turn to the requirements posed towards busi-
ness processes to define a quality standard. Based on the WfMC definition, the
foremost requirement for business processes is to realise a business objective.
We understand the business objective as achieving a state of the organisation
where pre-determind criteria are fulfilled – this will be discussed in detail in
Section 5. However, this is not yet sufficient as a quality standard: intuitively,
it is conceivable to have two business processes that both assuredly realise the
same business objective, but still divert widely in what we perceive as quality.
This is because we need to consider a second aspect as well: the term “business
process” implies an economic orientation or striving to maximise profits which,
in turn, means that business objectives should be realised as economically or
cost-efficiently as possible. Thus, a fundamental quality standard for business
processes can be defined as consuming minimum economic value to achieve a
given business objective. This results in the following definition:
Definition 1. Fundamental Business Process Quality: For a given busi-
ness objective, fundamental business process quality is defined as the relation
between minimum economic value consumed by an optimum business process to
achieve the business objective and actual economic value consumed by the busi-
ness process assessed to achieve the business objective.
Formally, fundamental business process quality can be expressed as follows:
fP Qualitybo :P → [0,1]
fP Qualitybo (P) := (evcmin,bo
evcact,bo(P)if achieved (bo)
0otherwise
evcact,bo (P)≥evcmin,bo >0⇒fP Qualitybo (P)∈[0,1]
where fP Qualitybo denotes fundamental business process quality with respect
to a business objective bo,Pdenotes a business process in the sense of a set
of process instances, Pdenotes the set of all business processes, evcmin,bo and
evcact,bo (P)denote minimum and actual economic value consumption to achieve
a business objective bo (see Section 5.2), and achieved (bo)denotes whether the
given business objective has been achieved (see Section 5.1.4).
In other words, optimum business process quality reflects the absence of waste
(i.e., fP Qualitybo assumes the optimum value of 1). Note that economic value
consumption is always positive according to the notion that there are no activi-
ties that do not use any resources3(see Section 3.1). Our fundamental definition
3We also require a direct causal link between business process and business objective,
so there are no business objectives that are achieved without a business process’s
contribution , i.e. without activities (see Section 5.1.1).
11
corresponds to the manufacturing-based view on quality in the sense that the
business objective is treated as a specification the process has to adhere to and
to the value-based view in the sense that we consider the economic value of re-
sources consumption. Note that fundamental business process quality according
to our definition always relates to a given business objective4. Altering the busi-
ness objective but a little always results in the need to re-assess business process
quality. This view is based on the WfMC definition of business processes which
implies a business objective as preliminarily granted. Nevertheless, we recognise
that irritations may be caused because, accordingly, a business process which
produces superior output with equal resources would not be considered as bet-
ter in quality. A short discussion on this topic is therefore included in Appendix
A. If a business process model is not capable of reaching the given business
objective in the first place, we can consume indefinite resources without ever
achieving the business objective, so business process quality will tend to zero.
Likewise, we do not consider “gradual” achievement of business objectives (see
Section 5.1.4). This means that if a business process in the sense of a set of pro-
cess instances does not achieve its business objective, its fundamental quality
will be zero as stated in our definition.
We now can match our fundamental definition against the requirements set out
in Section 2.
1. The measurability requirement leads to three main challenges for our fun-
damental definition. First, quality measurement presumes knowledge on the
resources consumption of a hypothetical optimum process with respect to
the business objective. Second, the economic value consumed by a business
process is hard to measure in practical settings. To illustrate this, consider
the valuation of upstream processes output. This may be done at costs of
a possibly sub-optimal process or at market rates that might be difficult to
obtain. Other examples are the cost for employing capital goods or the valu-
ation of risks incurred from non-compliance to public regulations. Third, the
dependency on a given concrete business objective impedes comparability
over time, between varying business processes and between organisations as
any change on these dimensions will lead to a change in the business objec-
tive; e.g., when considering changing output quantities required. Overall, we
can subsume that our fundamental business process quality definition will
not lend itself to concrete measurement in organisations.
2. The accountability requirement poses another issue because of the very in-
clusive nature of our fundamental definition. Consider the roles in a typical
organisation that will influence the fundamental quality of a business pro-
cess by setting business goals, designing processes, implementing processes,
providing process input or enacting processes. As all these tasks are typically
4According to the WfMC definition, we assume one business objective per process.
However, as we will see later, business objectives can be nested, i.e. aggregated
arbitrarily, so it is possible to combine aspired targets from different management
domains into one business objective for one business process.
12
carried out by different roles, but contribute to a common quality measure,
it will not be feasible to establish appropriate accountability.
3. Fulfilment of the transparency requirement is generally advanced by the sim-
ple nature of the fundamental quality definition. However, problems arise in
conjunction with our considerations on measurability: the complexity issues
that impede measurability also hinder transparency from an individual man-
ager’s perspective.
In summary, we can reach a fundamental definition of business process quality
in a straightforward and comprehensive manner. However, when matching this
definition against the requirements we set out, we must concede that its practical
applicability will be severely limited. To resolve this issue and foster real-world
relevance, we need to compromise on a fully comprehensive view on quality
considering two aspects:
–An approach based on comprehensive economic value assessment will be
too complex to be applicable in real-world settings, so we have to accept a
trade-off by identifiying “proxy” measures for relevant aspects of business
processes instead. These should be apt to reflect the fundamental quality
standard while fulfilling measurability and transparency requirements. For
instance, this might be achieved by assessing reasons for deviations from the
quality optimum.
–We need to consider various stages in a business process lifecycle to reflect the
involvement of different roles in an organisation and thus achieve account-
ability: the object of each quality measure should be within the domain of
control of a role in the organisaiton.
To reflect these conclusions, our approach will be to “decompose” our funda-
mental quality definition into relevant aspects of business processes that can be
subject to individual measures. These aspects will then be further analysed along
business process lifecycle stages that can be linked to accountable organisational
roles and provide a clearer notion on what can be considered as an optimum pro-
cess. Based on the ISO quality notion, we will then be able to define appropriate
quality standards for business process aspects within lifecycle stages as objects
of quality management. This will facilitate comparison to existing approaches
and identification of gaps where further research will be rewarding.
5 Quality-Related Aspects of Business Processes
As argued in Section 4, we aim to identify aspects of business processes that are
relevant to quality and can be subject to quality-related measures in practice.
Based on the components of our fundamental business process quality definition,
we thus enter into a more detailed discussion of business objectives and the
consumption of economic resources by business processes.
13
5.1 Business Objectives
According to the WfMC business process definition and in line with the utili-
tarian nature of business processes, each business process is linked to a business
objective - if there are activities within an organization that are not linked to a
business objective, they do not constitute a business process.
In this section, we first derive requirements we pose towards the definition of
valid business objectives for the purposes of quality assessment. We then discuss
existing approaches to business objectives, adapt the best suited approach to
reflect our requirements, and provide a formal definition of the achievement of
business objectives, which is a prerequisite for quality assessment.
5.1.1 Requirements for Valid Business Objectives
According to Definition 1, we need to establish whether or not a business ob-
jective is achieved as a prerequisite for quality measurement. A process in the
sense of a set of process instances that achieves its business objective can be
characterised as effective (the classic notion of “getting the right things done”
[38]) . We therefore require a clear understanding on how a business objective
must be made up and described to allow for its use as effectiveness criterion. In
this respect, we can derive requirements for valid business objectives that en-
able business process quality assessment from our fundamental business process
quality definition:
1. A business objective must be defined in a way that allows to clearly establish
whether or not the business objective has been achieved. A business objective
is to be understood from a comprehensive business model perspective and
corresponds to a desirable state of a sub-domain of the organisation. This
means that the business objective corresponds to a process in the sense of a
set of process instances, not to a single process instance. Business objectives
are thus always subject to cardinality and time constraints5– a real-world
business objective for a business process is achieved when a number of pro-
cess instances are successfully enacted within a certain timeframe. A single
successfully enacted process instance will usually not suffice to achieve a
business objective, but merely raise the number of successful instances by
one towards the required cardinality.
2. For a business process and its associated business objective, there must be a
direct causal relation between the contribution of the business process and
the achievement of the business objective.
5We can safely assume that there are no business objectives that do not comprise
a time constraint - in an extreme case, the time constraint would be the period of
operation of the organisation. Note that time constraints may be given in cyclic (i.e.
repetitive) or due date form. Due to the mostly repetitive nature of business processes
subject to systematic quality management, the former usually predominates.
14
The first requirement is instrumental. Note that, according to Definition 1, we
need to assess resource consumption caused to achieve the business objective. If
we cannot determine whether the business objective has been achieved, this is
not feasible. The second requirement is more relaxed than the WfMC business
process definition where a business process realises a business objective instead
of merely contributing to one. However, business objectives where more than
one business process contribute are also conceivable. In this case, the business
objective can easily be broken down to sub-objectives where each reflects just the
contribution of one process. It is also possible to have a business process that
contributes to multiple business objectives, but these can be combined into a
single one, as we will see in our adapted business objectives ontology. In any case,
a causal link needs to be established: if the business process’s causal contribution
towards its business goal cannot be established, it possibly constitutes a waste
of resources where no quality definition can be applied.
The second requirement entails that only operational business objectives that
can be fully controlled by the organisation are valid; i.e. strategic objectives
(e.g., in the sense of competitive positioning as in [39]) are exempt. To illustrate
the difference, consider the following example: A strategic objective for an organ-
isation might be to extend its market share. However, even if this goal is reached,
it is not possible to determine for any one process in the organisation whether its
execution was really required. For instance, the weakness of competitors might
have played a role as well. Moreover, one would be hard-pressed to describe a
business process where this could be given as the business objective according
to the WfMC definition. This is also reflected in the phrasing of Definition 1.
Example 2. As an example for valid business objectives, consider the helpdesk
process of an internet service provider:
–“Improve customer satisfaction” would not constitute a valid business objec-
tive because, without additional information, one cannot determine whether
or not the objective has been achieved. Additional information required com-
prises a definition of how customer satisfaction is measured as well as as-is
customer satisfaction.
–“Achieve 95% customer satisfaction” would not constitute a valid business
objective because no clear causal link to any business process can be estab-
lished. The level of customer satisfaction is a product of multiple business
processes as well as external factors – we have no way of knowing whether
95% customer satisfaction is due to an effective helpdesk process or to any-
thing else, so it does not make sense to use this as a determinant for helpdesk
process quality.
–“Solve all incoming customer tickets within two business days” or, more pre-
cisely, “All incoming customer tickets have been solved within two business
days” would constitute a valid business objective as it clearly describes a
measurable state of the organisation to be reached via the business process.
15
In our definition of fundamental business process quality, we also stipulated that
the business objective is given a priori, i.e., business process quality cannot be
improved by achieving a more desirable (or, higher quality) business objective.
For our helpdesk example, this means that a process that solves all customer
tickets within two days while consuming the same amount of resources is not
considered as of higher quality than the 95%-process. As this result may be irri-
tating at first, we reconsider the ISO quality definition: requirements cannot be
“over-achieved”. Moreover, treating the business objective as a factor in deter-
mining a process’s quality would preclude quality measurement of any sort. For
instance, delivering higher data rates might be even more desirable than 100%
ticket resolution: generally, there is no natural limit to optimising business objec-
tives, so assessing business process quality under consideration of the business
objective would incorporate comparison to a hypothetical optimum objective
that eludes analytic appraisal. For a more detailed discussion of this issue, refer
to Appendix A.
5.1.2 Related Work on Business Objectives
We are now able to discuss related work on business objectives in BPM based
on our requirements. Models for business objectives or goals6have been pro-
posed by Kueng and Kawalek [40], Neiger and Churilov [41,42], Soffer and Wand
[43], Markovic and Kowalkiewicz [44] as well as Lin and Sølvberg [45] based on
previous work [46]. All approaches implicitly or explicitly recognize differences
between strategic and business objectives. Results are summarised in Table 3.
As none of the approaches we discussed focuses on quality- or effectiveness-
related issues, it is not surprising that none fully matches our requirements. For
our purposes, we therefore adapt the approach of Lin and Sølvberg [45] which
has the additional advantage of being designed to accomodate the semantic needs
of distributed process models. This property might be useful when, as part of
future work, an attempt is made to automate business process quality evaluation
independently of single process modeling languages.
5.1.3 Adapted Business Objectives Ontology
The main topic we need to resolve results from the focus of Lin and Sølvberg’s
work on enabling collaboration in distributed process landscapes (e.g. cross-
organisational process choreographies) by annotating semantic content to pro-
cess models [46,45]. This implies that considerations are made on a modeling
instead of an enactment level. Accordingly, their ontology does not comprise a
concept for cardinality which will be required to satisfy our first requirement
6In the following, we only use the term business objectives while related work might
also refer to business goals. In the BPM space, these terms are generally used syn-
onymously.
16
Source Primary focus Strategic objectives Business objectives Requirement for
measurable definition
Requirement for causal
relation
Kueng and
Kawalek
Modeling business
processes based on goals
and evaluating business
process design
“Non-functional goals”
independent of any
single business processes
“Functional goals” used
as the basis for business
process modeling
Not fulfilled: no formal
measurable definition of
objectives
Fulfilled: business
processes are designed
based on objectives to
be achieved
Neiger and
Churilov
Applying the
value-focused thinking
approach [47] to
structure business
objectives for business
process modeling
“Strategic objectives”,
presumably
corresponding to
top-level objectives
within a hierarchy of
“fundamental
objectives”
“Functional or process
objectives” as means
objectives contributing
to fundamental
objectives of the
organisation
“Functional objective”
denotes the objective of
a function in the sense
of an atomic functional
node within an EPC
Not fulfilled: no
requirements posed
towards how a business
objective must be
specified
Partially fulfilled:
formal link between
processes and
objectives, but no
causal relation required
Soffer and
Wand
Incorporating business
processes’ contribution
towards “soft goals”
into business process
modeling by formalizing
concepts and
interrelations
“Strategic goals [...]
more abstract objectives
the organization is
striving to achieve”,
corresponding to “soft
goals” to evaluate
process alternatives
“Operational goals,
defining a desired state
to be reached by a
process”
Every process instance
terminates in a state
that belongs to the
process goal
Partially fulfilled: a
process goal is defined
as any state a respective
process instance can
terminate in. We
therefore can determine
when a process goal has
been achieved, but lack
the concept of a process
as a set of multiple
process instances with a
common objective
Fulfilled: the goal is
defined as the set of
states a process instance
can terminate in, “all
goal-defining state
variables are changed
from their initial value
at least once in at least
one path of the process”
Markovic
and Kowal-
kiewicz
Integrating business
goals into business
process modeling
“A strategic goal tends
to be longer term and
defined qualitatively”
“An operational goal is
a short-term
contribution to a
strategic goal”
Fulfilled: goals are
defined as “an
attainable, measurable
and time-bound state of
the world”
Partially fulfilled: goals
are formally linked to
business processes, but
the causal relation
between executing the
process and achieving
the goal is not
formalised
Lin and
Sølvberg
Developing a goal
ontology for semantic
annotation of
distributed process
models
“Soft goals can be
positively or negatively
satisfied by [..]
activities”
“Hard goals can be
achieved by [..]
activities”
Partially fulfilled: goals
are expressed with
respect to states of
activities or artifacts;
however, there is no
concept of multiple
process instances
contributing to a single
business objective
Fulfilled: goals are
explicitly linked to
activities that achieve
them
Table 3. Views on business objectives in BPM
17
towards valid business objectives. More specifically, we need to provide for the
notion of business objectives being achieved by sets of process instances instead
of individual process instances as discussed above. Additionally, we will need to
elaborate on the concept of constraints posed towards objectives.
To reflect the understanding of business processes set out in Section 3.1 and the
requirements discussed above, we adapt and extend the goal ontology proposed
by Lin and Sølvberg in [45] to a business objective ontology. To accomodate our
requirements, we undertake the following alterations:
–The term “goal” is replaced by “business objective” to be consistent to the
WfMC definition of business processes.
–A business objective reflects a desired state of the organisation. It is com-
posed of artifacts in desired states with respect to the values their attributes
assume and the satisfaction of additional constraints. This presumes that
attributes of artifacts are modeled in a way that they are apt to completely
describe desired states. This understanding is central to our approach. Gen-
erally, achieving a business objective will require more than one successful
process instance.
–Activities are omitted from the targets of business objectives because ex-
ecuting an activity does not constitute a target in itself from a business
perspective. It might rather be stipulated that a certain class of activities
must be executed to achieve a certain result which, in turn, is part of the
objective. In other words, the activity’s result, but not the activity in itself
is part of the business objective. We choose to model this as constraints
to the desired artifact states (see below). Consider the following example:
when handling incoming materials, it may initially constitute a goal to have
checked the materials for defects. From a business perspective, however, it
is not desirable to check materials, but to ensure that defects are detected
with a certain level of certainty (i.e. that the expected probability of a false
acceptance or “βerror” is below a certain threshold value). Whether this is
achieved by the concrete checking activity or by anything else does not mat-
ter: all activities that result in the required level of certainty are equivalent
from an objectives perspective. This can be modeled as a checked attribute
of the material artifact with a constraint of a given level of certainty.
–Similarly, constraints are omitted as direct target of business objectives. For
our purposes, business objective constraints generally relate to attributes of
artifacts, to the quantity of artifacts or to a timeframe, and are modeled as
such. If additional constraints are to be satisfied, they are therefore modeled
with reference to attributes.
–To achieve a more rigorous definition of business objectives, we replace the
unspecified relation between business objectives and (target) artifacts. In-
stead, our model explicitly represents artifacts’ residing in desired or unde-
sired states.
–We omit the concepts of soft versus hard goals as well as actor roles as these
are not central to our approach at this stage.
18
The resulting business objective ontology is depicted in Figure 2. The notation
used has been described in [48].
j
Business
Objective
X
Artifact
Target
X
Artifact
Target
X
Att ib t
g
Att
r
ib
u
t
e
Target
Timeframe Constraint
Cardinality
Att ib t
targetAttribute
desirableValues
Set
Symbols (not all possible combinations shown)
Explicit partial many-to-many relation
Class concept as basic construct
Attribute Type Artifact Type
Att
r
ib
u
t
e
Value
Implicit partial many-to-many relation
Explicit partial many-to-one relation
Each left-side set element explicitly relates to
exactly one right-side element
Artifact
X
Aggregation
Cartesian product
Fig. 2. Business objectives ontology
–A Business Objective is constituted by a non-empty set of Artifact Targets.
–An Artifact Target refers to exactly one Artifact Type. This relation is sub-
ject to a Cardinality constraint and a Timeframe constraint modeled as
attributes to the Artifact Target. Basically, the Artifact Target can be inter-
preted as part of the Business Objective that demands that a certain number
of Artifacts of a certain type are available in a certain state within a certain
Timeframe. As an example, consider a number of incoming payments that
should be posted within a week.
–As in the model proposed by Lin and Sølvberg, Business Objectives can
be nested. Following the WfMC definition which links each business process
to a business objective, we can easily aggregate and disaggregate business
processes in the same way as business objectives without violating the WfMC
or other common definitions. Therefore, it will be a subject of future work
to assess if there are any sensible rules on how this should be done or if the
structuring of Business Objectives can be done arbitrarily.
19
–The Cardinality of the Artifact Target should be specified as a relation and a
quantity, e.g. “> 5” while the quantity might also depend on other Business
Objectives, processes within a superordinate process chain or case occur-
rences, e.g. the number of incoming payments. If the Cardinality is based on
case occurrences, additional specification (e.g. of a related timeframe) is re-
quired. The Timeframe of the Artifact Target will in most cases be specified
in relative terms, e.g. “within one week”. As most processes subject to BPM
are repetitive in nature, absolute Timeframes like “by 31/12/2010” will be
the exception.
–Each Artifact has attributes according to Attribute Types defined via its
Artifact Type. We assume that there is at least an “ID” attribute we can
use to determine whether an Artifact exists, so there will be no Artifacts
without attributes. There is an implied one-to-many relation between Ar-
tifacts and Attribute Types. Each Attribute Type is associated to a set of
specific Attribute Values which represent the possible values an attribute of
the Attribute Type might assume.
–For each Artifact Target, there is a set of Attribute Targets. Attribute Tar-
gets related to a common Artifact Target may only relate to Attribute Types
associated with the Artifact Target’s Artifact Type. In other words, they
must be consistent to the Artifact Type of the Artifact Target. This is rep-
resented by an implicit relation between the Artifact Target-Attribute Target
and the Artifact Target-Artifact Type relations.
–Each Attribute Target is associated with a sub-set of the Attribute Values
of its Attribute Type. These represent the “desirable values” that must be
achieved to promote the Attribute Target and ultimately the Artifact Target
to a desired state.
–An Attribute Target may be associated with additional Constraints that
represent how the Attribute Target should be reached. As stated above, a
good example for this is the level of certainty associated with “checked”
attributes.
Note that the composition of business objectives from Artifact Targets readily
supports breaking down and combining business objectives as set out in Section
5.1.1 by splitting and joining their sets of Artifact Targets.
Example 3. Our business objective ontology can, for instance, be applied to the
helpdesk case from Example 2. In this case, our Business Objective is to solve all
incoming customer tickets within two days. This means that we need to consider
a single Artifact Target with “customer ticket” as its Artifact Type and a single
Attribute Target with “solution approved by customer” as its Attribute Type.
The single desirable Attribute Value is “true”. The sole challenge to model this
business objective lies in the Cardinality of its Artifact Target. Apparently, “all
incoming customer tickets” relates to case occurrences we need to specify further.
In this case, we can refer to the Timeframe of two business days. In practical
terms, it makes sense to collect all tickets of a business day as they relate to
a common absolute Timeframe of two business days later, so the number of
20
case occurrences within a business day provides the Cardinality of the business
objective. Note that we thus formally work with a different Business Objective
per business day! This issue will be subject to further discussion in future work.
5.1.4 Business Objective Achievement
In summary, we can now define when a Business Objective has been achieved:
Definition 2. Achievement of a Business Objective bo is defined as follows:
achieved (bo)⇔∀at ∈bo :
∃A|∀a∈A:Λa=Λat ∧ ∀att ∈AT Tat :vλatt
a∈AVatt∧(|A|relat cat)∧
satisfied (tat)
where
bo and at denote a Business Objective and an Artifact Target,
Aand adenote a set of Artifacts and a single Artifact,
Λaand Λat denote the Artifact Type of an Artifact aand the Artifact
Type associated with an Artifact Target at,
att and AT Tat denote an Attribute Target and the set of Attribute Targets
associated with an Artifact Target at,
λatt denotes the Attribute Type associated with an Attribute
Target att,
vλatt
adenotes the value of the attribute of an Artifact awith
Attribute Type λatt,
AVatt denotes the set of desirable Attribute Values associated with
an Attribute Target att,
relat and cat ∈Ndenote the relation associated with the Cardinality of an
Artifact Target at and the quantity associated with the
Cardinality of an Artifact Target at, while
satisfied (tat)signifies that the the timeframe constraint tat of an Artifact
Target at is satisfied.
As additional prerequisite, a Business Objective bo is only valid iff the following
holds:
∀att ∈AT Tat : (Λλatt =Λat ∧ ∀av ∈AVatt :λav =λatt)
where, in addition to the definitions stated above, Λλatt denotes the Artifact Type
associated with the Attribute Type λatt of Attribute Target att. In other words,
all Attribute Targets associated with an Artifact Target must (indirectly) refer to
the correct Artifact Type, and all Attribute Values associated with an Attribute
Target must refer to the correct Attribute Type.
21
Note that business objective achievement is defined binomially, i.e. there is no
grade to which an objective has been achieved, for instance in the sense of a
share of successfully completed transactions. This is due to the fact that gradual
achievement of business objectives would presume equal importance between all
Artifact Targets and between all individual Artifacts, which we cannot take as
granted generally.
We have now reached a measurable definition of when a business objective has
been achieved. For our second requirement with respect to valid business ob-
jectives, we refer to the integration of process models and the goal ontology
as described by Lin and Sølvberg, which clearly establishes a causal link and
satisfies our needs without further amendments.
We recognise that this rather mechanistic approach towards business objectives is
not sufficient to entirely capture soft goals as described in [43] as well as desirable
or undesirable “side effects” of business process execution, such as employees’
satisfaction, which might also be denominated as objectives for business pro-
cesses. In our approach, these effects need to be integrated into the Constraints
associated with Attribute Targets or the evaluation of resources consumption
as discussed in Section 5.2. However, a more practical approach to these issues
may be to accept that not all an organisation strives for is fit to be expressed
in terms of business processes. A discussion on related topics can be found in
Appendix A.
5.2 Economic Value Consumption
Having defined under which conditions a business objective can be considered as
having been achieved, we now look into the second component of Definition 1:
the consumption of economic value caused by a business process to achieve its
business objective. We first compare the concepts of economic value consumption
in business processes and business process input. We then examine the various
types of economic resources consumed by a business process as well as the cor-
responding unit costs. Lastly, we discuss optimum economic value consumption
and waste in business processes as a proxy measure to be applied instead of
comparing actual and optimum value consumption.
5.2.1 Economic Value Consumption versus Business Process Input
Business process input is one of the most widely-used terms in BPM, and is
also closely linked with the economic value consumed by business processes.
We therefore assess the scope of business process input versus the topics we
consider as valuable economic resources consumed in business processes. An
initial overview is given in Figure 3.
22
Economic Value
Consumption Business Process
Input
Capital goods
employment
Labour
Raw materials
Supplies
Non-consumables
Labour
Quasi-consumables
Fig. 3. Economic value consumption vs. business process input
The resources that constitute economic value consumption will be discussed in
more detail in the next section, including resources that are used and com-
pletely or partially consumed in business processes, but not typically considered
as business process input. Here, we only describe non-consumables as the part
of business process input we do not regard as constituent of economic value
consumption.
In our understanding, non-consumables comprise any tangible and intangible re-
sources that are utilised, but not consumed by a business process. As an example,
take information provided by an upstream process or plots of land. However, we
need to discern between “core” non-consumables and quasi-consumables. Quasi-
consumables are non-consumables which should be treated as consumables with
respect to economic value consumption. This occurs when non-consumables are
procured or produced (as output or business objective of upstream processes)
exclusively for utilisation in the business process under consideration. In this
case, the respective upstream process can be seen as part of the business process
in question, and procured non-consumables can be seen as equivalent to raw ma-
terials or supplies if they are used in individual process instances or as capital
goods if they are used for the process as a whole.
In the following sections, resources whose employment in business processes leads
to economic value consumption will be referred to as economic resources.
5.2.2 Types of Economic Resources in Business Processes
The total economic value consumption caused by a business process can be
decomposed as depicted in Figure 4. White boxes represent actual classes of
economic resources consumed by the process while grey boxes provide additional
levels on which these can be aggregated. The economic resources consumed in
business processes are presented in Table 4 with respect to their usage that in
turn leads to required quantities.
23
Economic Resource Description and Usage
Process Instances
Value Consumption
This category consists of resources that are consumed in
individual process instances; i.e. if there are no process
instances, the resources are not consumed. Note that resource
consumption need not be equal for each process instance, so the
number of process instances is only an approximate measure.
Materials Materials correspond to goods that are spent during the
execution of process instances. Mostly, this relates to tangible
goods, but we also need to treat some quasi-consumables in a
similar way. This occurs when information items are not spent
by a process instance, but when they have been created or
procured exclusively for the use of the process instance.
Raw materials Raw materials directly enter into the output of a business
process. Therefore, they are directly linked to the business
objective of a business process (e.g. if there is a bill of
materials) unless they are employed ineffectively. Sources of
waste consist of used raw materials that are not required
according to the business objective (“over-achieving” the
business objective) and rejects.
Supplies Supplies comprise materials that are employed and consumed
(or quasi-consumed) in a business process, but do not enter into
the business process output. As an example, consider power
usage by machinery during process execution.
Labour Cost is caused by personnel spending time to work on a process
or to attend to a process. Labour thus incorporates
characteristics of materials usage as well as capital goods
employment or even capital goods provision, as costs may arise
because personnel is kept ready, even if there is no work to be
executed on the process.
Capital Goods
Employment
Generally, the employment of tangible capital goods in process
instances causes costs. Intangible capital goods will cause costs
as they are provided (see below), but not as they are employed.
Use-bound
Depreciation
Tangible capital goods may deteriorate as they are used, i.e.
they may be subject to wear.
Maintenance The use of tangible capital goods may lead to the need to
execute ongoing maintenance processes which, in turn, causes
costs.
Capital Goods
Provision
Employing capital goods in business processes will cause costs
even if the resources are not used in process instances, but
simply as they are held ready for use. Therefore, these costs
arise independently of how many process instances are executed.
Time-bound
Depreciation
Capital goods may also deteriorate in value as time passes,
independently of their being used.
Capital Employment Holding ready capital goods will bind economic capital causing
opportunity costs.
Table 4. Types of economic resources
24
Economic Value
C
C
onsumption
Value of Capital
Goods Provision
Process
Instances Value
Consumption
Value of Capital
*
Value of
Value
of
Labour
Value of Time-
bound
Value of Capital
Goods
Employment
*
Materials
Value
of
Labour
bound
Depreciation
Goods
Employment
Value of Raw
Materials* Value of
Maintenance
Value of Use-
bound
Depreciation
Value of
Supplies*
* incl. quasi-consumables
Fig. 4. Economic resources tree
Note that according to the common use of the term in economics, raw materials
assume a special role in economic value consumption as they are directly “incor-
porated” into the business objective. This means that there is a natural lower
limit to the raw materials required to achieve a business objective. Transgress-
ing the natural lower limit of raw materials consumption is caused by process
instances that lead to refuse (i.e. the raw material consumed by the respective
instances is wasted) or a process design that deliberately uses excess raw ma-
terial per process instance e.g. for cutting scrap (i.e. each process instance uses
more raw material than strictly required due to process design).
Two additional aspects will be referred to future work: Process follow-up costs
are costs which are caused by process execution but occur later on, such as the
cost of non-compliance to external regulations. In many cases, follow-up costs are
related to risks incurred in the course of process exeuction. Follow-up costs are
akin to negative externalities in economics [49]. Sunk costs are costs that cannot
be avoided anymore, i.e. that occur regardless of whether or not a process is
actually executed. This is mostly the case if resources are made dedicated to a
process, such as capital goods that are procured and cannot be utilised otherwise.
5.2.3 Unit Costs
To determine the economic value consumed, we also require unit costs. We refer
to unit cost in the economic sense; i.e., unit costs describe appropriate economic
costs that may or may not correspond to price. In this respect, we need to
25
consider several issues. Generally, resources might be valued at actual costs or
at opportunity costs. Opportunity costs describe the value that is lost because
resources are not employed otherwise, e.g. used in other processes that may
generate additional economic value. Opportunity costs are very similar to alter-
native business objectives and are therefore not considered in our context (see
Appendix A) – we can assume zero elasticity of supply with respect to resources
employed, which means that resources can be procured or produced as required
without change in price or costs. However, the actual cost of resources must
also be subject to closer discussion as it is affected by upstream processes (e.g.
in procurement or the choice of location for labour). We defer this question to
later work. For capital goods, depreciation is traditionally valued via acquisition
costs and useful life (mostly external accounting, see for instance IAS 16 [50]) or
replacement costs and useful life (internal accounting). Both pose similar issues
as actual costs for other resources.
The determination of unit costs is a major driver of complexity for business
process quality assessment as differing methods and therefore differing results
lead to shifts in the relative weighing of resources with respect to their influence
on overall business process quality. As example consider the use of capital goods
such as information systems enabling process automation. In a high labour cost
environment, it may be an economically sound decision to spend capital on
additional process automation while this may not be sensible anymore if process
execution is relocated e.g. to an offshore low cost environment.
5.2.4 Optimum Economic Value Consumption and Proxy Measures
According to Definition 1, we will not only require an understanding of the eco-
nomic value of resources consumed by a business process to assess its quality,
but also of optimum (i.e. minimum) value consumption required in case of an
optimum process to achieve the process’s business objective. In the following, we
refer to a business process that is effective with respect to a given business objec-
tive and optimal with respect to its economic value consumption as a comparable
optimum process. According to Definition 1, a common business objective is the
only limitation that needs to be observed to consider a hypothetical business
process as comparable to an actual one.
Even for simple business objectives, it will in practice be very difficult, if not
impossible, to determine minimum economic value consumption with a compa-
rable optimum process. The first reason for this is that it is difficult to evaluate
economic resource consumption for a comparable optimum process as well as for
an actual process. The second reason is that neither the comparable optimum
process design nor the comparable optimum settings for process execution are
known. This issue is exacerbated with added “degrees of freedom” with respect
to process design and alterations to the execution environment of the business
process that might be made to obtain a comparable optimum process. Process
design determines the activities to be executed as part of the process and thus
26
the economic resources required. The process execution environment comprises
external suppliers, upstream business processes, availability of capital goods,
and unit labour costs (e.g. via the choice of location). It thus determines unit
costs for economic resources. Each dimension will need to align to the other one
in order to sustain an optimum balance. For instance, if the process execution
environment is changed to a low labour cost location, process design may have
to be realigned to substitute the use of capital goods with labour. It becomes
clear that there is no singular optimum process design and no singular optimum
process execution environment – both dimensions depend on each other.
As already touched upon in Section 4, in this respect our approach will be to as-
sess appropriate scopes of influence or domains for differing BPM lifecycle stages
in the course of future work. This will not only provide us with clear organisa-
tional responsibilities, but also enable us to at least partially resolve the degrees
of freedom issue by constricting the frame of action available on each stage. Ba-
sically, the limitations to be observed in the definition of a comparable optimum
process become more severe as we progress in the BPM lifecycle because the de-
grees of freedom available diminuish. In particular for the lower-level stages, this
will lead to examining sources of economic waste that can be allocated to the
respective stage as a proxy measure from deviations from a comparable optimum
process. For example, on the lifecycle stage for actual process execution, there
might be deviations from the optimum in the form of resource waste that could
be avoided by more diligent work. However, resource unit costs are generally not
in the responsibility of the respective organisational roles.
6 Related Work
We already touched upon major related work with respect to basic aspects of
QM and BPM in Section 3. Existing approaches on business objectives as a
prerequisite for business process quality assessment were discussed in Section
5.1.2.
There is some related work that deals with the quality of business process mod-
els: van der Aalst introduced soundness of Workflow Nets [51], Weber and Re-
ichert developed process model refactoring [52], Li and Reichert addressed ref-
erence model discovery by model merging [53], Becker et al. discussed process
model quality focusing on certain stakeholder groups and applications, Guce-
gioglu and Demirors applied software quality characteristics to business pro-
cesses [54], Mendling assessed formal errors in EPC business process models
[55], Cardoso analysed workflow complexity as one possible measure for process
model quality [56], and Vanderfeesten et al. discussed quality metrics in business
process modeling [57,58]. As all of these approaches are related to business pro-
cess design, we will refer to them in our future discussion of quality on differing
BPM lifecycle stages. However, we need to be aware that these approaches are
all based on individual quality drivers or characteristics. They are not formally
27
derived from a comprehensive definition of business process quality, but rather
aimed at covering single aspects of business process model quality in detail. Our
approach intends to complement this work with a comprehensive framework for
business process quality as a whole.
A more comprehensive attempt to develop a “Quality of Business Processes
(QoBP) framework” focusing on process modeling was made by Heravizadeh et
al. [59]. Business process quality is defined in terms of 41 quality dimensions
which are derived from related work, e.g. in the field of software engineering.
Insofar, the approach lists quality characteristics, but does not provide a singular
definition of quality which would, for instance, enable quality measurement. This
also means that there is no way to clearly establish why each quality dimension
is important, in which way it contributes to overall quality or if the listed quality
dimensions are sufficient to explain overall quality. In our approach, we intend
to address these issues by first providing a well-founded fundamental quality
definition, and then deriving contributive characteristics in a second step. The
QoBP approach has been presented in more detail in [60]. In this context, quality
has been defined as non-functional but distinguishing characteristics of a business
process. We do not concur with that view as we define fundamental business
process quality under consideration of a business objective to be achieved. In our
view, this better suits the ISO quality definition. However, we concede that non-
functional characteristics are difficult if not impossible to evaluate in economic
terms, so the two approaches to quality can be seen as complementary.
Heinrich and Paech also made a proposal to define the quality of business pro-
cesses mostly based on insights from the field of software quality [61]. While this
approach lists various quality characteristics, which may provide guidance to
our future work, it does not integrate them into a comprehensive formal quality
definition. Therefore, it is not yet apt to fulfil our measurability requirement.
Moreover, we stipulate that the adoption of software engineering results to busi-
ness process problems still requires closer analysis.
Business process reengineering and optimisation constitutes an area which, tak-
ing our value-based fundamental definition, is closely related to optimising busi-
ness process quality. [19,18] provide good examples for the “classic” all-encompassing
reengineering view. Reengineering approaches mostly comprise recommended
best practices and other informal methods which are mostly based on anec-
dotal evidence (see the sample case in Appendix B). [62,63,64,65] and, with a
focus on well-defined process models, [66] constitute additional good examples
for optimisation based on informal methods. This view is also reflected in the
OMG Business Process Maturity Model [67] and other BPM maturity models
[68] which suggest criteria to allocate business processes to maturity levels with-
out giving clear evidence on how this structure is devised. While this informal
character fits well with practical applicability on the one hand, we still lack an
overarching comprehensive model to ensure causal relations between measures
recommended and intended results as well as completeness of coverage of qual-
ity or optimisation aspects. From our perspective, this may be addressed by
28
structuring measures along our fundamental definition and BPM lifecycle stages
which will be attempted in future work. The other class of related optimisation
approaches focuses on the analysis of formalised business processes or workflows.
Examples include [69,51], where Petri nets are proposed to leverage existing anal-
ysis methods, and [70], where various optimisation strategies for process designs
with given input and output sets per activity are discussed. These approaches
are mainly suited to optimising control flow and resource scheduling, as they
typically do not address individual activities in terms of necessity, effort or pos-
sible alternatives. They thus constitute important tools but are not sufficient to
guarantee optimum business process design. We intend further analysis to be
part of future work on quality in the process design lifecycle stage.
Process performance management and business activity monitoring constitute
another related area linked to the quality of process execution. Research in this
area is very much driven by practical requirements, and we will refer to it in a
future discussion on quality on the process execution lifecycle stage. Exemplary
work includes [71,72] and reflects the close association of this research issue
to industry and tool vendors. Also in the field of process execution, Grigori
et al. have developed a proposal to monitor and manage exceptions in process
execution [73]. We will refer to this in our later research on the respective BPM
lifecycle stage.
7 Conclusion
In this paper, we specified requirements with respect to the practical applicability
of business process quality measures. We discussed existing views on business
processes and approaches to quality. We derived a fundamental business process
quality definition that satisfies both fields, but will require further detailing to
achieve practical applicability. However, this constitutes a framework to guide
further research and to appraise related approaches – which are mostly focused
on particular aspects – in a broader context. The components of our fundamental
definition were assessed in more detail to identify areas where additional work
will be required. This provides us with a sound basis from which we can extend
our work in the field of quality management for business processes.
Future research will focus on business process quality measures that are founded
on our fundamental definition from Section 4 and fulfil the requirements set out
in Section 2. As set out above, a major issue in this respect relates to assessing
various BPM lifecycle stages regarding business process quality. Other fields to be
examined comprise the identification of resources waste or optimum comparable
processes as well as practical evaluation of economic value consumption. Process
follow-up costs and sunk costs have already been mentioned as exemplary for
this field. We also intend to analyse how business processes and business ob-
jectives should be structured to enable effective quality assessment. The reason
for this is that, based on common business process definitions, process models
29
may be arbitrarily aggregated or decomposed, e.g. to obtain well-manageable
process documentation. We intend to analyse how the impact of this on quality
assessment results can be eliminated or at least minimised. Defining business
objective cardinalities in case of recurring transactional processes has already
been mentioned in this respect (see Example 3). Finally, the integration of QM
in BPM support tools will be an important issue for practical application.
References
1. Phillips, L.W., Chang, D.R., Buzzell, R.D.: Product quality, cost position and
business performance: A test of some key hypotheses. Journal of Marketing 47(2)
(Spring 1983) 26–43
2. Capon, N., Farley, J.U., Hoenig, S.: Determinants of financial performance: a
meta-analysis. Management Science 36(10) (October 1990) 1143–1159
3. Forker, L.B., Vickery, S.K., Droge, C.L.: The contribution of quality to busi-
ness performance. International Journal of Operations & Production Management
16(8) (1996) 44–62
4. Adam, E.E., Corbett, L.M., Flores, B.E., Harrison, N.J., Lee, T., Rho, B.H., Rib-
era, J., Samson, D., Westbrook, R.: An international study of quality improvement
approach and firm performance. International Journal of Operations & Production
Management 17(9) (1997) 842–873
5. Dale, B.G., van der Wiele, T., van Iwaarden, J., eds.: Managing Quality. 5th edn.
Wiley-Blackwell (2007)
6. Parasuraman, A., Zeithaml, V.A., Berry, L.L.: A conceptual model of service
quality and its implications for future research. Journal of Marketing 49(4) (1985)
41–50
7. Harvey, J.: Service quality: a tutorial. Journal of Operations Management 16(5)
(1998) 583–597
8. Porter, M.E.: Competitive Advantage: Creating and Sustaining Superior Perfor-
mance. Free Press, New York (1985)
9. Miller, J.G., Vollmann, T.E.: The hidden factory. Harvard Business Review 63(5)
(September-October 1985) 142–150
10. Emore, J., Ness, J.: The slow pace of meaningful change in cost systems. Journal
of Cost Management 5(4) (1991) 36–45
11. Küting, K., Lorson, P.: Grenzplankostenrechnung versus Prozeßkostenrechnung
(in German). Der Betriebs-Berater 21 (1991) 1421–1433
12. Deming, W.E.: The new economics: for industry, government, education. 2nd edn.
MIT Press (2000)
13. Sousa, R., Voss, C.A.: Quality management re-visited: a reflective review and
agenda for future research. Journal of Operations Management 20(1) (2002) 91–
101
14. Crosby, P.B.: Quality is free: the art of making quality certain. New American
Library (1980)
15. International Accouting Standards Board: Framework for the
Preparation and Presentation of Financial Statements (1989)
http://eifrs.iasb.org/eifrs/bnstandards/en/framework.pdf, accessed
on 13/04/2010.
30
16. Davenport, T.J., Short, J.E.: The new industrial engineering: Information tech-
nology and business process redesign. Sloan Management Review (Summer 1990)
11–27
17. Hammer, M.: Reengineering work: don’t automate, obliterate. Harvard Business
Review (July-August 1990) 104–112
18. Davenport, T.J.: Process innovation: reengineering work through information tech-
nology. Harvard Business School Press (1993)
19. Hammer, M., Champy, J.: Reengineering the corporation. A manifesto for business
revolution. HarperBusiness (1993)
20. Lindsay, A., Downs, D., Lunn, K.: Business processes – attempts to find a defini-
tion. Information and Software Technology 45(15) (December 2003) 1015–1019
21. van der Aalst, W.M.P., Berens, P.J.S.: Beyond workflow management: product-
driven case handling. In: Proceedings of the 2001 International ACM SIGGROUP
Conference on Supporting Group Work, ACM (2001) 42–51
22. Workflow Management Coalition: Workflow management coalition terminology &
glossary 3.0 (February 1999) http://www.wfmc.org, document reference WFMC-
TC-1011, accessed on 03/11/2009.
23. Weske, M.: Business Process Management. Springer (2007)
24. Dale, B.G.: The Received Wisdom on TQM. In: Managing Quality. 5th edn.
Wiley-Blackwell (2007) 58–73
25. International Organization for Standardization: ISO 9000:2005: Quality manage-
ment systems – Fundamentals and vocabulary (2005)
26. Guler, I., Guillén, M.F., Macpherson, J.M.: Global competition, institutions, and
the diffusion of organizational practices: The international spread of ISO 9000
quality certificates. Administrative Science Quarterly 47(2) (June 2002) 207–232
27. Neumayer, J., Perkins, R.: Uneven geographies of organizational practice: explain-
ing the cross-national transfer and diffusion of ISO 9000. Economic Geography
81(3) (2005) 237–259
28. International Organization for Standardization: The ISO Survey - 2007 (2008)
http://www.iso.org/iso/survey2007.pdf, accessed on 05/04/2010.
29. Garvin, D.A.: What does "product quality" really mean? Sloan Management
Review 26(1) (Fall 1984) 25–43
30. Garvin, D.A.: Competing on the eight dimensions of quality. Harvard Business
Review 65(6) (1987) 101–109
31. Sower, V.E., Fair, F.K.: There is more to quality than continuous improvement:
listening to Plato. Quality Management Journal 12(1) (2005) 8–20
32. van der Aalst, W.M.P., ter Hofstede, A.H.M., Weske, M.: Business process man-
agement: A survey. In: International Conference on Business Process Management
2003. Volume 2678/2003 of Lecture Notes in Computer Science., Springer (2003)
33. Christensen, C.M., Rosenbloom, R.S.: Explaining the attacker’s advantage: tech-
nological paradigms, organizational dynamics, and the value network. Research
Policy 24(2) (March 1995) 233–257
34. Bower, J.L., Christensen, C.M.: Disruptive technologies: catching the wave. Har-
vard Business Review 73(1) (January-February 1995) 43–53
35. Reeves, C.A., Bednar, D.A.: Defining quality: alternatives and implications.
Academy of Management Review 19(3) (1994) 419–445
36. Leffler, K.B.: Ambiguous changes in product quality. American Economic Review
72(5) (December 1982) 956–967
37. Feigenbaum, A.V.: Quality control: Principles, practice, and administration.
McGraw-Hill, New York (1951)
31
38. Drucker, P.F.: The effective executive. Harper & Row (1967)
39. Porter, M.E.: What is strategy? Harvard Business Review November-December
(1996) 61–78
40. Kueng, P., Kawalek, P.: Goal-based business process models: creation and evalu-
ation. Business Process Management Journal 3(1) (1997) 17–38
41. Neiger, D., Churilov, L.: Structuring business objectives: a business process mod-
eling perspective. In: International Conference on Business Process Management
2003. (2003)
42. Reijers, H.A., Vanderfeesten, I.T.: Cohesion and coupling metrics for workflow
process design. In: International Conference on Business Process Management
2004. Volume 3080/2004 of Lecture Notes in Computer Science., Springer (2004)
43. Soffer, P., Wand, Y.: On the notion of soft-goals in business process modeling.
Business Process Management Journal 11(6) (2005) 663–679
44. Markovic, I., Kowalkiewicz, M.: Linking business goals to process models in se-
mantic business process modeling. In: International IEEE Enterprise Distributed
Object Computing Conference 2008. (2008)
45. Lin, Y., Sølvberg, A.: Goal annotation of process models for semantic enrichment
of process knowledge. In: International Conference on Advanced Information Sys-
tems Engineering 2007. Volume 4495/2007 of Lecture Notes in Computer Science.,
Springer (2007) 355–369
46. Lin, Y., Strasunskas, D., Hakkarainen, S., Krogstie, J., Sølvberg, A.: Seman-
tic annotation framework to manage semantic heterogeneity of process models.
In: International Conference on Advanced Information Systems Engineering 2006.
Volume 4001/2006 of LNCS., Springer (2006) 433–446
47. Keeney, R.L.: Creativity in decision-making with value-focused thinking. Sloan
Management Review 35(4) (Summer 1994) 33–41
48. Sølvberg, A.: Data and what they refer to. In: Conceptual Modeling. Volume
1565/1999 of Lecture Notes in Computer Science. Springer (1999) 211–226
49. The Economist: Economics A-Z, Keyword: Externality
http://www.economist.com/research/economics/alphabetic.cfm?letter=E,
accessed on 01/04/2010.
50. International Accouting Standards Board: International Ac-
counting Standard 16: Property, Plant and Equipment (2003)
http://eifrs.iasb.org/eifrs/bnstandards/en/ias16.pdf, accessed on
13/04/2010.
51. van der Aalst, W.M.P.: The application of Petri nets to workflow management.
The Journal of Circuits, Systems and Computers 8(1998) 21–26
52. Weber, B., Reichert, M.: Refactoring process models in large process repositories.
In: International Conference on Advanced Information Systems Engineering 2008.
Number 5074 in Lecture Notes in Computer Science (2008) 124–139
53. Li, C., Reichert, M., Wombacher, A.: Discovering reference models by mining pro-
cess variants using a heuristic approach. In: International Conference on Business
Process Management 2009. Volume 5701 of Lecture Notes in Computer Science.
(2009) 344–362
54. Guceglioglu, A.S., Demirors, O.: Using software quality characteristics to measure
business process quality. In: International Conference on Business Process Man-
agement 2005. Volume 3649/2005 of Lecture Notes in Computer Science., Springer
(2005) 374–379
55. Mendling, J.: Detection and Prediction of Errors in EPC Business Process Models.
PhD thesis, WU Wien (2007)
32
56. Cardoso, J.: Business process quality metrics: Log-based complexity of workflow
patterns. In: On the Move to Meaningful Internet Systems 2007. Volume 4803/2010
of Lecture Notes in Computer Science., Springer Berlin / Heidelberg (2007) 427–
434
57. Vanderfeesten, I., Cardoso, J., Mendling, J., Reijers, H.A., van der Aalst, W.:
Quality metrics for business process models. In: Workflow Handbook 2007. Future
Strategies (2007) 179–190
58. Vanderfeesten, I., Reijers, H., van der Aalst, W.M.: Evaluating workflow process
designs using cohesion and coupling metrics. Computers in Industry 59(5) (May
2008) 420–437
59. Heravizadeh, M., Mendling, J., Rosemann, M.: Dimensions of business processes
quality (QoBP). In: Business Process Management Workshops 2008. Volume 17 of
Lecture Notes in Business Information Processing. (2009) 80–91
60. Heravizadeh, M.: Quality-aware Business Process Management. PhD thesis,
Queensland University of Technology (2009)
61. Heinrich, R., Paech, B.: Defining the quality of business processes. In: Modellierung
2010. Volume P-161 of Lecture Notes in Informatics. (2010) 133–148
62. Speck, M., Schnetgöke, N.: To-be Modeling and Process Optimization. In: Process
Management: A Guide for The Design of Business Processes. Springer (2003) 135–
163
63. Reijers, H.A., Liman Mansar, S.: Best practices in business process redesign: an
overview and qualitative evaluation of successful redesign heuristics. Omega 33(4)
(August 2005) 283–306
64. Liman Mansar, S., Reijers, H.A.: Best practices in business process redesign: use
and impact. Business Process Management Journal 13(2) (2007) 193–213
65. Jansen-Vullers, M., Kleingeld, P., Loosschilder, M., Netjes, M., Reijers, H.: Trade-
offs in the performance of workflows – quantifying the impact of best practices.
In: Business Process Management Workshops 2007. Volume 4928/2008 of Lecture
Notes in Computer Science. (2007) 108–119
66. Becker, J., Rosemann, M., von Uthmann, C.: Guidelines of Business Process Model-
ing. In: Business Process Management: Models, Techniques, and Empirical Studies.
Volume 1806/2000 of Lecture Notes in Computer Science. Springer (2000) 241–262
67. The Object Management Group: Business process maturity model (BPMM) (June
2008) http://www.omg.org/spec/BPMM/1.0/PDF/, accessed on 09/10/2009.
68. Smith, H., Fingar, P.: Process management maturity models (July 2004)
69. van der Aalst, W.M., van Hee, K.M.: Business process redesign: A Petri-net-based
approach. Computers in Industry 29 (1996) 15–26
70. Hofacker, I., Vetschera, R.: Algorithmical approaches to business process design.
Computers and Operations Research 28(13) (November 2001) 1253–1275
71. Castellanos, M., Casati, F., Dayal, U., Shan, M.C.: A comprehensive and auto-
mated approach to intelligent business processes execution analysis. Distributed
and Parallel Databases 16(3) (November 2004) 239–273
72. IDS Scheer: Process intelligence white paper: What is process intelligence? (Jan-
uary 2009) http://www.process-intelligence.com, accessed on 07/04/2010.
73. Grigori, D., Casati, F., Dayal, U., Shan, M.C.: Improving business process quality
through exception understanding, prediction, and prevention. In: Proceedings of
the 27th International Conference on Very Large Data Bases 2001, Morgan Kauf-
mann (2001) 159–168
74. Cao, G., Clarke, S., Lehaney, B.: A critique of BPR from a holistic perspective.
Business process management journal 7(4) (2001)
33
75. Hammer, M.: The process audit. Harvard Business Review (April 2007) 111–123
76. Thommen, J.P.: Gabler Wirtschaftslexikon, Keyword: Wirtschaftlichkeitsprinzip
(2010) http://wirtschaftslexikon.gabler.de/Archiv/8349/wirtschaftlichkeitsprinzip-v5.html,
accessed on 05/01/2010, in German.
A Business Objectives as Drivers of Business Process
Quality?
In our fundamental business process quality definition, we stipulated that busi-
ness process quality can only be established with respect to a given business
objective. Moreover, we only need to determine whether a business objective has
been reached, not how sound, important or desirable the business objective is for
the organisation. This, in turn, means that optimising the business objective will
not lead to superior business process quality, or that we can denote a business
process’s quality as “good” even if we do not accept the business objective pur-
sued as sensible – an assertion that might seem surprising at first. To obtain a
better understanding, we investigate three topics in this respect: quality views as
discussed in Section 3.2, views on business process design and optimisation and
implications from treating business objectives as variable for quality assessment.
A.1 Quality Views with Respect to Business Objective Optimization
The view that business objectives do not constitute an object for quality assess-
ment corresponds well with the manufacturing-based approach to quality and
with the ISO quality definition, which stipulates that requirements are to be
achieved but does not challenge the requirements themselves. With respect to
the other quality views, we must conclude that the role of business objectives
in our fundamental quality definition does not fit too well with them. In the
following, we will therefore need to substantiate why we preclude alternative
interpretations.
A.2 Business Objectives as an Object for Business Process Design
According to our definition, setting effective business objectives is not part of
business process quality management, and the quality of business objectives must
be judged independently on the quality of their objectives. A similar question
refers to whether business objectives are treated as an object or as a constraint
of business process design. While this question has not been focussed directly by
authors on business processes, we nevertheless can identify views on the issue
that have evolved since the early days of business process reengineering.
During the early 1990s, authors such as Davenport and Short or Hammer and
Champy advocated business process reengineering as a radical way of changing
34
organisations. This involved reconsidering the objectives that were pursued by
enacting business processes7. One demand made was that all processes (or, more
precisely, process objectives) should be oriented towards “the customer” which
reflects an user-based quality view (see Section 3.2). Of course, “customer” was
a rather flexible notion as departments began to define other departments or
management as their customer. However, it is clear that, at that time, assessing
business processes involved assessing the business objectives that were pursued
as well. If asked to ascertain the quality of a business process, we can assume
that the advocates of reengineering would have commenced with scrutinising the
associated business objective.
The reengineering approach, however, led to a number of issues that could not be
easily resolved, such as change management and generally people-related topics
[74]. One of these issues was that there was no clear way to support decisions on
which processes serve the customer and which do not, which business objectives
are important and which are not etc. Overall, it became apparent that radical
approaches towards business process optimisation that did not focus on impro-
ving given processes but involved reconsidering whether process objectives made
sense at all quickly led to a huge scope of change that overtaxed most if not all
organisations. Consequently, the question which processes are actually required
and, inherently, wich business objectives are to be pursued or if one business
objective is to be preferred over another has been omitted from more recent ap-
proaches towards BPM. This trend even pertains to recent work by Hammer as
one of the founders of the process reeingineering management philosophy, which
can be observed in [75].
As a result, the mentioned decisions have been delegated to strategic manage-
ment so that the business objectives associated with business processes are not
the object of BPM anymore. This corresponds to the insight that business ob-
jectives are difficult to analytically assess with currently available methods, so
it makes sense to treat them as part of a discipline not as analytically focu-
sed as BPM. For instance, Porter’s influential article “What is Strategy?” refers
to operational effectiveness as the ability to “get more out of their inputs than
others because they eliminated wasted effort, employ more advanced technology,
motivate employees better or have better insight into managing particular ac-
tivities or sets of activities” [39]. A large part of these topics relates to BPM.
Competitive strategy, however, is defined as “deliberately choosing a different
set of activities to deliver a unique mix of value”. Trade-offs between differing
business objectives are thus seen as the subject of strategic management rather
than BPM.
7See, for instance, [18, p. 10]: “Process innovation [...] involves stepping back from a
process to inquire into its overall business objective”, or [19, p. 35]: “In doing reen-
gineering, businesspeople must ask the most basic questions about their companies
and how they operate: Why do we do what we do? And why do we do it the way
we do?”
35
In summary, we conclude that our approach to exclude business objectives from
the drivers of business process quality is broadly in line with current thinking
on the scope of business process management as a whole.
A.3 Implications from Treating Business Objectives as Variables
for Quality Assessment
Treating the definition of business objectives as a factor for the quality of the
respective business process would lead to a quality standard comprising both op-
timum business objectives and minimum economic value consumption, instead
of just minimum economic value consumption with achieving the business objec-
tive as a side condition. These alternative quality standards converge with two
of the modes of the classic economic principle [76]: the minimum principle as
minimising input for a given amount of output (corresponding to Definition 1)
and the general optimum principle of reconciling input and output to the best
economic advantage of the organisation 8.
Accordingly, the classic issue of the general optimum principle would also apply
to a quality standard comprising business objectives as well as resources con-
sumption. For primitive cases like maximising the amount of products manufac-
tured while minimising material employed, it is clear that the general optimum
principle cannot be pursued. When a more sophisticated phrasing like balanc-
ing input and output as to maximise economic value generation is chosen, this
general issue can be resolved, but two other obstacles remain: the requirement
of valuating business process output, and the lack of comparability.
Evaluating the quality of business objectives in appropriate economic terms in-
volves providing an economic evaluation of the business process output aspired
as part of the business objective. Compared to evaluating resource consumption
by business processes (see Section 5.2), this is far more complex. This can be
illustrated by the example of an outgoing payments process. Business process
input in this case consists of the checked invoices to be paid, the information
system employed and labor. Even if additional assumptions for instance with
respect to depreciation of the information system is required, we can assume
that economic evaluation can be conducted at cost (see Section 5.2.3). On the
output side, assessing the economic value of having executed payments would
involve analysing what happens when each individual payment is not made in
terms of economic cost – an area where we clearly leave sound analytic ground
and enter into speculation.
Another issue arising with evaluating quality including business objectives refers
to the lack of comparability. If we include the business objective in our quality
standard, we not only need to establish whether economic value has been wasted
8The maximum principle as maximising output for a given amount of input is also
conceivable as a quality standard for business processes, but leads to similar issues
as the general optimum principle.
36
on resources, but at the same time whether this value could have been put to
better use or how much value could have been generated by any other process.
In other words, this means that the system of comparable alternative processes
that had been established via a common business objective disintegrates. The
quality standard becomes undefinable because of its lack of boundaries, so quality
assessment of a concrete process is not possible anymore.
Overall, we can conclude that optimum business objectives as part of the quality
standard would certainly drive the level of complexity that needs to be resolved
beyond practical limits, thereby compromising the measurability and transpa-
rency requirements.
A.4 Conclusion
As discussed in the previous paragraphs, the view that good business objectives
constitute part of business process quality corresponds well to most views on
quality, but not to the manufacturing-based view and the derived ISO quality
definition. Also, scrutinising business objectives as part of BPM is not in line
with current thinking on the scope of BPM. The decision which business ob-
jectives should be pursued is rather seen as a subject of business strategy or
the definition of the organisation’s business model. However, the biggest issue
precluding variable business objectives in business process quality assessment
are the implications with respect to practical applicability. We therefore omit
optimum business objectives from our business process quality standard.
B The Credit Note Procedure Example
To illustrate the case for “radical” business process reengineering, Davenport
and Short as well as Hammer and Champy in articles that were both published
in summer 1990 described an example for process optimisation at Ford Motor
Corporation [16,17]. The example relates to optimisation of the accounts payable
process, one of the best-understood processes in terms of optimisation potentials
in administrative functions.
The authors claim that Ford’s North American operations were able to reduce
capacity requirements in the accounts payable department by 75% by radically
reengineering the process: instead of receiving and checking invoices, credit notes
are issued to suppliers upon the receipt of goods. It is illustrative to take another
look at the topic today.
Almost 20 years later, one would expect a practice that has been implemen-
ted with as much success at a well-known multinational company to have gai-
ned wide-spread acceptance. Personal observations during the last years, mainly
when working with European manufacturing groups as a consultant, show that
37
this is indeed the case and the practice is well-known and adopted in many com-
panies. However, it is by no means pervasive. The practice is, for instance, very
prevalent at automotive OEMs (“original equipment manufacturers”, i.e. car ma-
kers), but not widely spread in the machine tools industry. Overall, it could be
argued that as an estimate, on average less than 10% of total purchasing volume
are processed via a credit note procedure.
This outcome can be ascribed to the fact that by applying credit note procedures
instead of receiving invoices, work is not “obliterated” as claimed by Hammer,
but merely shifted from the customer to the supplier: instead of the customer
checking the invoice, the supplier checks the credit note. The new activities on the
supplier side involve matching the original customer order against the delivery
note and the credit note, which is rather similar to the original invoice checking
process. This, of course, is only possible in industries where buyers are in a good
bargaining position, hence the wide-spread adoption by automotive OEMs. As
opposed to the claim of the advocates of business process reengineering, the
workload has not been obliterated but merely reassigned. Moreover, the pressure
to adopt a credit note procedure has lessened with the advent of advanced process
automation techniques in the field, such as EDI and early scanning.
The example shows that even in prominent cases, the business process reengi-
neering postulation of radically re-thinking whether activities or processes really
need to be executed not always leads to tremendous results in the long run. Most
reengineering successes that in reality led to more than incremental results were
based on a combination of process reengineering, organisational reallocation of
tasks and elimination of fringe activities.
38
