A NEW FUZZY MODEL FOR EVALUATION OF KEY PERFORMANCE INDICATORS AND

PROCESS QUALITY

Danijela Tadić, Miladin Stefanović, Slavko Arsovski, Snežana Nestić


Abstract. Improvement of key business processes is one of requests of standard ISO 9001:2008 and has a
critical effect on the competitive advantage of any organization. In this paper key business processes, their
objectives, key performance outcomes and key objective indicator for each identified objective are defined
with respects to critical success factors. Modelling of the relative importance of purchasing key performance
indicators and their values, as well as modelling of influence of specific objective on process quality will be
presented in this paper using fuzzy sets. The algorithm for evaluation of KPIs and process quality is presented
and illustrated by example of purchasing process in one service organization.
Keywords: process quality strategy, effectiveness, fuzzy set, fuzzy AHP, fuzzy logic

1. INTRODUCTION

Quality management includes all the activities that organizations use to direct, control, and coordinate
quality. These activities include formulating a quality policy and setting quality objectives. They also include
quality planning, quality control, quality assurance, and quality improvement. According to literature review
(Deming, 1993 Juan, 1988, Crosby, 1984) it could be stated that achievement of quality goals leads to
improvement of the competiveness, effectiveness and flexibility of an organization. This is a reason why
considered problem has become a topic of research for both industry and academy in last decades. The
quality goals and objectives could be considered as part of strategic goals and objectives. In other words the
formulation of quality goals and quality strategy are based on the strategy development process and proposed
in different researches (Kaplan, Norton, 2008). The quality goals are defined by top managers with respect to
vision (how the organization wants to be perceived by the world), mission (what organization wants to be
achieve) and values (prescribe its behavior, character and culture). In the further steps top managers define a
goals and critical success factors (CSFs)- which the organization must accomplish to achieve the mission, by
examination and categorization of the impact (Oakland, 2004). These CFS are sufficient for the mission to be
achieved and to use for identification of the key business process in an organization. Business Process
Management (BPM) defines objectives of key business processes with respects to CFSs which must be
accompanied by measurable of key performance indicators (KPIs Improvement of key business processes
could be achieved if objectives can be measured through key performance indicators (KPIs). KPIs, on the
other hand, are measures that quantify management objectives, along with a target or threshold, and
enable the measurement of strategic performance. In the literature, there are numerous process
performance measurement methods (Neely et al, 1995). In these measurement systems values of variables are
expresses by precise numbers.
Performance measurement systems are developed on different mathematical and logical frameworks. With
respects to recommendation of ISO 9001:2008 that each enterprise should develop approach in improvement
of processes the issue of definition of KPIs definition of their weights, importance and values, as well as their
influence of the quality of specific process could be stated as important issue.
Considering the fact that enterprises exist in rapid changing environment, it is not easy to evaluate the
variables with precise numbers. In this paper, authors developed a new approach for assessment of quality of
specific process and weights and values of KPIs using fuzzy sets. It is assumed that decision makers of BPM
express their judgments far better by using linguistic expressions than by representing them in terms of
precise numbers. Uncertainties in the relative importance and values of measures of objectives caused by lack
of appropriate information and conflicting evidence (some of the information available is wrong, information
of non-relevant features of the considered problem, etc) are described by linguistic expressions. There are
numerous theories which claim to be the only proper tool to model linguistic expressions. According to
Zimmermann (Zimmermann, 1997), fuzzy sets theory can be the most appropriate way for modelling
linguistic expressions.
In this paper we will address following issues: modelling of the relative importance of purchasing key
performance indicators and their values, as well as modelling of influence of specific objective on process
quality. The algorithm based on fuzzy sets will be presented that will enable evaluation of each KPI, as well
as their importance on specific objective and further evaluation of process according to realization of each
objective.
The paper is organized in the following way. The section 2 describes the problem statement. Modelling of
uncertainties are described in Section 3. A new fuzzy model based on fuzzy logic for evaluation operational
strategy execution effectiveness is proposed in Section 4. In Section 5, a proposed model is illustrated by
example with real-life data. Conclusions are presented in Section 6.

2. LITERATURE REWIEV

Purchasing deals primarily with managing all aspects related to the inputs for an organization (i.e., purchased
goods, materials, and services), while supply chain management deals with inputs, conversion, and outputs.
As a result of implementation of SCM in manufacturing companies, the quality of the purchasing process is
now playing a key role in management of the selection of an optimal supply process, which is strategically
crucial (Erol, 2009). Having a high level of quality of the purchasing process is very important for all
companies, especially for manufacturing companies where more than 65% financial resources is spent on the
purchasing of resources. There are different aspects of ISO 9001’s implementation in terms of the
relationships with three key supply chain (SC) management practices (internal processes, supplier
relationships, and customer relationships). Prajogo et al. (2012) indicates that supplier and internal process
management both have a positive effect on operational performance, while customer process management
has no significant impact. The quality level of the purchasing process would, directly, affect the quality level
of the supply chain (Wu and Wang, 2011). Therefore, improving the level of quality of the purchasing
process must be significant (Sánchez-Rodríguez 2003, Sánchez-Rodríguez and Martínez-Lorente, 2004;
Paulraj et al., 2006, Soroush (2012)).) for enhancing the quality level of the supply chain. Establishing target
performance levels and then implementing metrics to track performance, relative to those targets, gives the
data necessary to continually improve performances.
The relation between implementation of quality management systems such as ISO 9001 (ISO 9001:2008) and
the quality of processes and outcomes is clear and has been identified in many researches. Obtaining a clear
understanding of business process quality constitutes the most important prerequisite (Lohrmann and
Manfred, 2013). Koc (2007) provides results showing that ISO 9000’s implementation makes a significant
difference to a firm’s performance when comparing certified and non-certified firms. There are various ways
in which a company can claim that its QMS meets the requirements of ISO 9001. These include certification
and party assessment or even self-assessment. In assurance of quality, self - assessment has an important
place. Rezaie et al. (2011) introduced a self - assessment system based on the fundamental principles and
requirements of ISO 9004:2000 and provided a tool for assessment of organizational maturity and recognition
of improvement points. A number of researches have also focused on the application of different approaches
in the selection, ranking and assessment of parameters (including suppliers) or even management systems.
For instance, Tsai and Choua (2009) proposed a novel hybrid model for selecting optimal management
systems under resource constraints. Flegel and Brozova (2011) presented a fuzzy decision model for
selection of an appropriate management system (ISO 9001 or ISO 14001). Feng (2004) presented a method of
fuzzy integrated estimation of the effectiveness of a quality management system. An improved fuzzy model
for evaluation of the effectiveness of QMS was presented in Liu et al. (2008). Identifying quality
improvement opportunities in a enterprises is not an easy task (Fore, 2011).
One of the first steps in improvement is identification and assessment of weak spots and self-assessment. The
assessment model could identify impacts on the existing system as well as provide the result of introducing
new standards Khir and Kai (2012). The assessment of the quality of processes provides a platform to
compare and benchmark different purchasing processes and indicators, as well as define actions for their
improvement. There are different solutions for QMS assessment and steering (MATLAB's GUI components
and its Fuzzy Logic Toolbox (Lee et al., 2011) or the Pareto Analytical-Hierarchy Process (PAHP) and the
Multichoice Goal Programming (MCGP) (Mahmoud et al., 2011)).
Psomas et al. (2013) developed an instrument that measures the effectiveness of the ISO 9001 Quality
Management System (QMS), based on its components, meaning the ISO 9001 objectives. In that research
ISO 9001 objectives and their indicators are identified from the literature, and the focus was on SMEs. The
Performance measurement can be achieved through: Key Result Indicators – KRIs, Result Indicators - RIs,
Performance Indicators – PIs and Key Performance Indicators - KPIs (Parmenter, 2010). KPIs are focused on
aspects that are critical to the success of the company. Their purpose is to enable performance measurement
and evaluation of processes and organization. They also indicate the quality of the processes, activities and
the company. Proper selection of KPIs is essential for effective performance measurement because too many
KPIs as well too complex KPIs can cause unnecessary waste of time and money.
Traditionally, the monitoring of a purchasing department's performance is primarily focused on cost analysis
and the evaluation of suppliers rather than on internal service aspects Holschbach and Hofmann (2011), so
some researches have focused on the internal service quality of purchasing departments (Large and König,
2009). Performance and quality measurement is an essential element of effective planning and control as
well as decision making. The measurement results reveal the effects of strategies and potential opportunities
(Bhagwat, 2007). It is clear that very important issues is measurement of the quality of specific process. Each
process could be measured according their objectives and accompanied KPIs. It is very important to provide
formal model for assessment the KPIs, their values as well as the influence of specific objectives on quality
of process. This is important for companies because it provides a platform to find weak spots, and provides
improvement actions, comparing different purchasing processes.


2. PROBLEM STATEMENT

In this Section, problem statement of definition of efficiency of quality of process is given which is placed in
the internal perspective of strategic map.
According to CSFs, strategic management defines key processes which could be decomposed on sub-
processes. The sub-processes under each business process are presented by set indices P { } P ,.., p ,..., 1 = . For
each sub-process of treated a key business process defines objective. In this paper, objective of a sub-
processes of any key business process are formally presented by set indices I { } I ,.., i ,..., 1 = where i is index of
objective and I is total number of objectives of considered business sub process. Each objective i, i=1,…,I is
associated with number of KPIs which are presented by set J { }
i
J ,.., j ,..., 1 = . The index j denotes KPI of
objective i and the total number of KPIs associated objective i, i=1,..,I is denoted as
i
J . The values of KPIs
for objective i, I ,.., 1 i = have different type (benefit type and cost type) and their values could be presented in
different units, for example monetary unit, time unit, percent, quantity, index, etc.
The relative importance of sub-processes under considered key business process are not equal and they are
determined by BMP team. It is realistic to assume that relative importance of KPIs are not equal. They are
assessed by each decision maker of BMP team. The relative importance of each pair of KPIs are stated by
fuzzy pair-wise comparison matrix. The elements of this matrix are triangular fuzzy numbers
|
|
.
|

\
|
=
i
'
jj
i
'
jj
i
'
jj
i
'
jj
~
u , m , l ; x W I ,.., 1 i ; J ,..., 1 j , j
i
'
= =

The normalized weights vector of KPIs is denoted as
| |
i
xJ 1
ji i
w W = . The relative importance of objective, | |
xI 1
i
w W = is calculated by using similar way.

The target of each identified process and KPOs targets can be determined by using different measurement
methods (survey, interview method, evaluation of leadership team, internal and/or external benchmarking,
etc). It can be stated that target of KPIs , objectives and process are denoted as I ,.., 1 i ; J ,.., 1 j t , t , t
i p i ji
= =
respectively. The target vales could be measured in percents, i.e. belong to interval [0-1].

The current values of KPIs are given by using different measurement procedure. The overall value of each
KPI is based on evaluation of BMP. BMP make decision with respect to target value and current value. In
decision making, decision makers, beside two values connected with each KPI, respect a KPI type. With
resects the fact that KPIs have measurement units, by using presented procedure for determining of overall
value of KPIs, they become non dimensional values which could be compared. The value of each KPI of
each objective are described by predefined linguistic expressions with respect its current and target values.
Process quality strategy effectiveness is determined by using proposed fuzzy algorithm. According to these
values BMP team makes decision about redesigning of quality strategy.

3. MODELLING OF KPIs RELATIVE IMPORTANCE AND VALUES FOR PURCHASING
PROCESS

3.1 Basic definition of fuzzy sets
Fuzzy sets are represented by its membership function which the parameters are shape, granularity and
location on the universe of discourse. The membership function shape of a fuzzy set can be obtained based on
one’s experience, subjective belief of decision makers, intuition and contextual knowledge about the concept
modeled (Zimmermann, 1978), uncertainty available on the treated linguistic variables (Berman, Trubatch
1997). However, subjectively in determining membership function has been considered as the weakest point
in fuzzy sets theory. Among the commonly used fuzzy numbers, triangular fuzzy numbers offer a good
compromise between descriptive power and computational simplicity. The triangular fuzzy numbers are
applicable to the present study. Fuzzy sets of higher types and levels have not as yet played a significant role
in applications of fuzzy sets theory (Klir, Yuan, 1995). Granularity is defined as number of fuzzy numbers
assigned to the relative importance parameters, their values, and level of effectiveness Lootsma (Lootsma,
1997) suggested that only seven categories at most can be used. The domain of fuzzy sets can be defined on
different measurement scales.
Some definitions of fuzzy sets in (Dubois, Prade, 1979, Klir Folger, 1988, Zimmermann, 2001) presented
used notations and are reviewed.
Definition 1. Uncertainty implies that in a certain situation a person does not dispose about information
which quantitatively and qualitatively is appropriate to describe, prescribe or predict deterministically and
numerically a system, its behavior or other characteristic (Zimmeramnn, 2001).
Definition 2. Linguistic variables are variables whose values are not numbers but words or sentences in a
natural or artificial language (Zadeh, 1975).
Definition 3. Fuzzy set
~
A is defined as a set of organized pairs:
( ) ( )
)
`
¹
¹
´
¦
s µ s e µ = 1 x 0 , X x x , x A
~ ~
A A
~
(2.1)
where:
Fuzzy set
~
A is defined on the universe set XeR. In general, set X can be either finite or infinite. ( ) x
A
~
µ is a
membership function of fuzzy set
~
A.
Definition 4. A fuzzy number
~
A is a convex normalized fuzzy set
~
Aof the real line R such that: (1)if exist
R x e
0
such that ( ) 1
0
~
= x
A
µ and (2) ) (
~
x
A
µ is piecewise continuous.
Definition 5. Fuzzy number
~
A on R is to be a triangular fuzzy number if its membership function ( ):
~
x
A
µ
R | | 1 , 0 ÷ is equal to
( )
| |
| |
¦
¦
¦
)
¦
¦
¦
`
¹
¦
¦
¦
¹
¦
¦
¦
´
¦
e
÷
÷
e
÷
÷
= µ
otherwise 0
u , m x
u m
u x
m , l x
l m
l x
x
~
A
(2.2)
Where u m l s s , l and u stand for the lower and upper value of the support of X respectively, and m for the
modal value. The triangular fuzzy number can be denoted by (l, m, u). The support of X is the set of elements
{ } u x l R x < < e . When l=m=u, it is a non-fuzzy number by convention.

Definition 6. The operations of fuzzy numbers are based on the theorem set by Dubois and Prade (Dubois,
Prade, 1980). Let two fuzzy numbers ( )
¦
)
¦
`
¹
¦
¹
¦
´
¦
e µ = R x | x
~
, x A
A
~
and ( )
¦
)
¦
`
¹
¦
¹
¦
´
¦
e µ = R y | y
~
, y B
B
~
. The membership functions
of these fuzzy numbers are subjective from zero to one and * is a continuous binary operation. Then
~ ~
B * A is
a fuzzy number which is denoted
~ ~ ~
B * A C = , such as
~
C . Values in domain of fuzzy number
~
C, can be
calculated as z=x*y and ( ) ( ) ( )
|
|
.
|

\
|
µ µ = µ
=
y
~
, x
~
min sup z
~
B A
y * x z
C
.
Definition 7. Defuzzifacation is an operation which determines the scalar or the crisp value that is the best
representative of a fuzzy set.

3.2 Modelling of the relative importance of purchasing key performance indicators

All KPIs are usually not of the same relative importance. It can be assumed that, their importance are
unchangeable during the considered period of time. We think that the judgment of each pair of treated KPIs
best suits human-decision nature (by analogy with AHP method). In other words, the relative importance of
KPIs of each objective i is stated by pair-wise comparison matrix. The elements of these matrix are
judgemental by decision makers of BPM which use predefined linguistic expressions. BMP team make
decision by consensus. They are modelled by and described by triangular fuzzy numbers,
|
|
.
|

\
|
=
i
jj
u ,
i
jj
m ,
i
jj
l ; x
i
jj
~
W
' ' '
'
I ,.., 1 i ;
i
J ,..., 1
'
j , j = = . The domains of defined triangular fuzzy numbers are defined
on standard scale measures [1-5]. The value 1 means that relative importance of KPI j and KPI
'
j of
objective i, have equal the relative importance. The value 5 means that relative importance of KPI j over
KPI
'
j of objective i, I ,.., 1 i ; J ,.., 1 j , j
i
'
= =

has most important.
These triangular fuzzy numbers which are given in the following way:
- low importance- ( ) 3 , 1 , 1 ; x
1
~
R =
- moderate importance- ( ) 5 , 2 , 1 ; x
2
~
R =
- high importance- ( ) 5 , 5 , 3 ; x
3
~
R =
If high relative importance of KPI
'
j over KPI j of objective i, I ,.., 1 i ; J ,.., 1 j , j
i
'
= = holds, then pair-wise
comparison scale can be represented by the fuzzy number
|
|
|
|
.
|

\
|
=
|
|
|
.
|

\
|
i
jj
l
1
,
i
jj
m
1
,
i
jj
u
1
; x
i
jj
~
W
' ' '
'
.
Weights vector of the considered KPIs of objective i,
n
I ,.., 1 i = is calculated by applying the method in
(Chang, 1996). The normalized weights vector
i
xJ 1
i
j
i
j
w W
(
¸
(

¸

= .
i
j
W is a non-fuzzy number and this gives the priority weights of one KPO over the other under objective i,
i
J ,.., 1 j I ,.., 1 i = = .

3.3 Modelling of purchasing key performance indicator values

In general, the KPIs of each objective can be different benefit type or cost-type and they have different
measurement units, for instance percent, days, etc. Each KPI are associated two values: target value and
current value.
BPM uses predefined linguistic expressions for describing KPI values. These linguistic expressions are
modelled by triangular fuzzy numbers, whose domains belong to the common scale [0-1]. The value 0
denotes very low value of KPI. The value 1 indicates a very high value of KPI. The triangular fuzzy numbers
for modelling the KPO values are:
- very low value- ( ) 2 . 0 , 0 , 0 ; y
1
~
v =
- low value- ( ) 5 . 0 , 3 . 0 , 1 . 0 ; y
2
~
v =
- medium value- ( ) 7 . 0 , 5 . 0 , 3 . 0 ; y
3
~
v =
- high value- ( ) 9 . 0 , 7 . 0 , 5 . 0
4
~
v =
- very high value- ( ) 1 , 1 , 8 . 0 ; y
5
~
v =


3.4 Modelling of importance of specific objective on purchasing process quality
In this paper, influence of specific objective on process quality is modelled by one of five predefined
linguistic terms. These linguistic expressions are modelled by triangular fuzzy numbers which domains
belong to interval [0-1]. The value 1 denotes total lack of gap between objective value and its target. The
value 0 indices a very high value gap between objective value and its target. Granularity of these triangular
fuzzy numbers is performed by respect to opinions of decision makers.
The triangular fuzzy numbers for modelling imporantance of specific objective on process quality:
- no important - =
1
~
s ( ) 2 . 0 , 0 , 0 ; y
-
low importance-
( ) 5 . 0 , 0 , 0 ; y s
2
~
=
- medium importance- ( ) 8 . 0 , 6 . 0 . 4 . 0 ; y s
3
~
=
- very important - ( ) 1 , 1 , 7 . 0 ; y s
4
~
=
-
highly importnat-
( ) 1 , 1 , 9 . 0 ; y s
5
~
=

4. THE PROPOSHED ALGORITHM FOR ASSESMENT OF KPIs AND PURCHASING PROCESS
QUALITY

For the BPM carrying outs the analysis, the following tasks are important: (1) to estimate how much each
KPI of treated objective could contribute to closing the value gap, and (2) to evaluate effectiveness of process
quality strategy used for redesign of strategy.
The Algorithm of proposed fuzzy model is presented as follows:

Step 1. Input pair-wise fuzzy comparison matrix of the relative importance of KPI for each objective i:
i
xJ J
i
jj
~
i
~
J ,.., 1 j ; I ,.., 1 i , W W
i i
'
= =
(
(
¸
(

¸

=

By using extent analysis which is proposed in (Chang, 1996), weights vector of KPIs at the objective level, is
calculated.

Step 4. Calculate weighted normalized KPI values for each objective i,
ji
~
V :
( )
|
|
|
.
|

\
|
µ =
|
|
.
|

\
|
= y ; y v V
ji
w
~
ji
v
ji
w
ji
~
ji
~
, I ,..., 1 i ; J ,.., 1 j
i
= =
The membership function values of the fuzzy number
ji
~
V

are calculated by using fuzzy operation dilatation
(Zimmermann, 2001, Dubois, Prade, 1980).

Step 5. Define the objective weight i:
i
ji
~
i
J ,.., 1 j
i
~
J ,.., 1 j ; I ,.., 1 i , V D = = =
=


Step 6. The overall weighted value of objective i is given:
I ,.., 1 i , D d
i
w
i
~
i
~
=
|
|
.
|

\
|
=

Step7. Calculate representative scalar fuzzy number
i
i
~
d , d by method of maximal possibility (Dubois, Prade,
1980).
i
d =defuzz I ,.., 1 i d
i
~
=

Step 8. Determine value gap of objective i:

I ,.., 1 i ,
ti
d
g
i
i
= =
The process quality effectiveness in observed objective i, i=1,...I can be defined according to the rule of fuzzy
logic (Zimmerman, 2001, Klir, Folger, 1988):
IF the value gap
i
g , THAN strategy efficiency is described by linguistic expression where
( )
*
k
~
s
i
i
~
s
3 ,.., 1 k
g y max µ = = µ
=
.
Step 9. Calculate weighted value of treated process

¿
=
· =
I
1 i
i p
d
I
1
d
Step 10. Determine value gap for process p:
P ,.., 1 p ,
tp
d
g
p
p
= =

The process quality effectiveness in observed key process p, p=1,...P can be defined according to the rule of
fuzzy logic (Zimmerman, 2001, Klir, Folger, 1988):
IF the value gap
p
g , THAN strategy efficiency is described by linguistic expression where
( )
*
k
~
s
p
i
~
s
3 ,.., 1 k
g y max µ = = µ
=
.

5. ILUSTRATTIVE EXAMPLE – EVALUATION OF PURCHASING PROCESS

The proposed model is tested on the real-life data which are obtained from service organization from central
Serbia. The selected process I purchasing process which is divided into seven sub-processes: define
purchasing requirements, purchasing planning, suppliers of evaluation, contracting, verification of the
purchasing, reclamation to supplier, and contract implementation monitoring.
The purchasing process is not very precise so, therefore, the decomposition of the purchasing process varies
from company to company. In some companies, more attention is paid to planning of the purchasing, in
others the selection of supplier, and in some companies’ attention is on both. Every company finds the most
suitable way, which depends on its size, structure, types of activities and development of communication.
In view of the growing supply of reference models (such as CIMOSA, GRAI-GIM and PERA, and others), in
various domains and in different formats, the potential benefits to organizations using them (Spiegel and
Caulliraux, 2012), and the different approaches in the decomposition of the purchasing process, (Johnson and
Chia-Yen, 2012) we selected a solution that could be suitable for small and medium sized manufacturing
companies.
Many authors have presented a similar model of the purchasing process. In this paper the purchasing process
is decomposed by using Structural system analysis (SSA) which is one of the process approach methods. The
purchasing process is decomposed to the levels that are necessary for effective management of quality and
process.
In this paper the 7 most common purchasing sub-processes in the manufacturing companies are selected
based on the author’s research and experience at the Center for Quality, Faculty of Engineering in Kragujevac
(Fig.1.):
- Definition of purchasing requirements,
- Purchasing planning,
- Suppliers’ evaluation,
- Contracting,
- The purchasing goods verification,
- Reclamation to supplier and
- Contract implementation monitoring.
The target and current value of defined KPI are presented in Table 1.
Table 1 Objectives, KPIs of identified objectives of purchasing process, target and measured KPI values
Objectives of
specific sub -
processes
KPI Description
Target Measured
data
The
normalized
value
Filling of
purchasing
requirements
(i=1)
Accuracy
(j=1)
Denied requests’ percentage based on
accuracy
0% 45%
3
~
v
Time (j=2)
Average time required to provide
purchasing decisions, calculated from
the date of receipt of the request
requirements
5 days 20 days
1
~
v
Level of
purchasing
planning (i=2)
Percentage of
purchasing
plan
realization
(j=3)
Total purchasing contracts’ value in
purchasing plan/Purchasing planning
value
100% 80%
4
~
v
Percentage of
realization
(j=4)
Contract realization value/Value of
contract realization
100% 75%
3
~
v
Completeness
(j=5)
Purchasing percentage value that
requires a change in the approved
purchasing plan
< 5% 6%
2
~
v
Inventories
level (j=6)
Inventories value purchased in the
reporting period/Inventories value
purchased in the previous reporting
period (x100)
≤ 85% 110%
1
~
v
Level of suppliers
evaluation (i=3)
Deadline
(j=1)
Delayed deliveries percentage
(0-1)% 4%
2
~
v
Quality
verification
(j=2)
Deliveries’ percentage with the
inconsistencies established for the
qualitative acceptance
1% 0%
3
~
v
Supplier
approval
status (j=3)
Consequences of nonconformities
3% (0-1)%
1
~
v
Level of
contracting effect
(i=4)
Time (j=1)
Average time required for the
contract provision, calculated from
the date of providing purchasing
decision
<3 days 13 days
1
~
v
Completeness
(j=2)
Percentage of changed tender
documents after the announcement
0% 2%
3
~
v
Procedure
correctness
(j=3)
Percentage of accepted complaints of
potential suppliers
0% 1%
3
~
v
Partnership
(j=4)
Comparison of real prices and
catalogue prices
100% 97%
5
~
v
Level of
verification of the
purchasing goods
(i=5)
Success (j=1)
Inconsistencies percentage of
detected external control
0% 1%
4
~
v
Level of
reclamation to
supplier (i=6)
Success
(j=1)
Successful reclamation percentage of
the total complaints number after the
identified nonconformities in
exploitation
100% 90%
5
~
v
Level of contract
implementation
monitoring (i=7)
Success (j=1) Contracts percentage realized on time
100% 75%
3
~
v
The relative importance of the purchasing sub-process is given by pair-wise comparison matrix:
(
(
(
(
(
(
(
(
(
(
(
(
(
¸
(

¸

1 , 1 , 1 1 , 1 , 1 1 , 1 , 1 R / 1 1 , 1 , 1 R / 1 1 , 1 , 1
1 , 1 , 1 1 , 1 , 1 R R / 1 1 , 1 , 1 R / 1 R / 1
1 , 1 , 1 R / 1 1 , 1 , 1 R / 1 R / 1 R / 1 1 , 1 , 1
R R R 1 , 1 , 1 R R / 1 R
1 , 1 , 1 1 , 1 , 1 R R / 1 1 , 1 , 1 R / 1 1 , 1 , 1
R R R R R 1 , 1 , 1 R
1 , 1 , 1 1 , 1 , 1 R R / 1 1 , 1 , 1 R / 1 1 , 1 , 1
1
~
1
~
1
~
1
~
1
~
1
~
1
~
2
~
1
~
3
~
1
~
1
~
2
~
1
~
1
~
2
~
1
~
1
~
2
~
1
~
1
~
3
~
1
~
2
~
3
~
1
~
2
~
3
~

By applying procedure which is proposed in (Chang, 1996), the weights vector of purchasing sub-processes is
calculated. It can be assumed that the objective weights are equal of the purchasing sub-processes weights
and they are:
06 . 0
7
w , 109 . 0
6
w , 043 . 0
5
w , 268 . 0
4
w , 105 . 0
3
w , 314 . 0
2
w , 101 . 0
1
w = = = = = = =
The relative importance of KPIs under each identified objective is given by pair-wise comparison matrix. The
weights of KPIs are calculated by fuzzy extended analysis (Chang, 1996).
The relative importance of KPIs of filling of purchasing requirements (i=1):
(
¸
(

¸

1 , 1 , 1 1 , 1 , 1
1 , 1 , 1 1 , 1 , 1

5 . 0
12
w , 5 . 0
11
w = =
The relative importance of KPIs of level of purchasing planning (i=2):
(
(
(
(
(
(
(
¸
(

¸

1 , 1 , 1
1
~
R / 1
2
~
R / 1
3
~
R / 1
1
~
R 1 , 1 , 1
1
~
R / 1
2
~
R / 1
2
~
R
1
~
R 1 , 1 , 1
1
~
R / 1
3
~
R
2
~
R
1
~
R 1 , 1 , 1

04 . 0
24
w , 194 . 0
23
w , 337 . 0
22
w , 431 . 0
21
w = = = =
The relative importance of KPIs of level of suppliers evaluation (i=3):
(
(
(
(
(
(
¸
(

¸

1 , 1 , 1
2
~
R / 1
1
~
R / 1
2
~
R 1 , 1 , 1 1 , 1 , 1
1
~
R 1 , 1 , 1 1 , 1 , 1

201 . 0
33
w , 455 . 0
32
w , 343 . 0
31
w = = =
The relative importance of KPIs of level of contracting effect (i=4):
(
(
(
(
¸
(

¸

1 , 1 , 1 1 , 1 , 1 1 , 1 , 1 1 , 1 , 1
1 , 1 , 1
1 , 1 , 1 1 , 1 , 1 1 , 1 , 1
1 , 1 , 1 1 , 1 , 1 1 , 1 , 1 1 , 1 , 1
1 , 1 , 1 1 , 1 , 1 1 , 1 , 1 1 , 1 , 1

25 . 0
44
w , 25 . 0
43
w , 25 . 0
42
w , 25 . 0
41
w = = = =
The relative importance of KPI for the rest objectives are: 1
71
w , 1
61
w , 1
51
w = = = .
The weighted values of KPIs of level of contracting effect (i=4) is presented in Fig. 1.

Fig. 1 The weighted KPI s values of level of contracting effect
The weighted value of level of contracting effect (i=4) is presented in Fig. 2

Fig. 2 The weighted value of of level of contracting effect
The weighted values of each identified objective of purchasing process are calculated by applying Algorithm
(from Step 1 to Step 6).
By using Algorithm (Step 7 to Step 8) the weighted values of purchasing process objectives, the gap values
for each objective and effeteness of quality strategy are calculated and presented in Table 2.
Table 2 The weighted objective values, gap values and quality strategy effectiveness

i
d
i
t for the first
period
i
g The quality effectiveness
i=1 0.3361 0.6 0.5602 medium importance
i=2 0.4206 0.8 0.5257 medium importance
i=3 0.4264 0.9 0.4737 medium importance
i=4 0.4935 0.8 0.6169 medium importance
i=5 0.7 0.8 0.875 very important
i=6 0.9281 1 0.9281 highly important

i=7 0.5744 0.8 0.7180 medium importance
By using IF-THAN rules of fuzzy logic, we can determine quality effectiveness for each identified objective
of purchasing process.
( ) 801 . 0 5602 . 0 g max : 1 i
3
~
s
=
|
|
.
|

\
|
= µ = , ( ) 6285 . 0 5602 . 0 g max : 2 i
3
~
s
=
|
|
.
|

\
|
= µ =

( ) ( ) ( ) 3685 . 0 3685 . 0 , 0526 . 0 max 4737 . 0 g , 4737 . 0 g max : 3 i
3
~
s
2
~
s
= =
|
|
.
|

\
|
= µ = µ =

( ) 9155 . 0 6169 . 0 g max : 4 i
3
~
s
=
|
|
.
|

\
|
= µ =

( ) 5833 . 0 875 . 0 g max : 5 i
4
~
s
=
|
|
.
|

\
|
= µ =

( ) 282 . 0 9281 . 0 g max : 6 i
5
~
s
=
|
|
.
|

\
|
= µ =

( ) ( ) ( ) 41 . 0 06 . 0 , 41 . 0 max 7180 . 0 g , 7180 . 0 g max : 7 i
4
~
s
3
~
s
= =
|
|
.
|

\
|
= µ = µ =
For specific illustrative example it is possible to conclude that sub-process i=4 (Level of contracting effect) is
the most influential on quality of purchasing process. In the other words using the presented algorithm it is
possible to conclude to which extent specific sub-process could contribute to the quality of process (in this
case purchasing process).
The weighted value of purchasing process is given by using express (Step 9 of the proposed Algorithm), so
that 6710 . 0 d
p
= . Let us the target value of considered process is 0.72. The value gap for purchasing process
is given by Algorithm (Step 9):
9319 . 0
72 . 0
6710 . 0
g = =
IF the value gap 9319 . 0 g = , THAN process quality effectiveness is described by linguistic expression
( ) ( ) ( ) 7731 . 0 319 . 0 , 7731 . 0 max 9319 . 0 y , 9319 . 0 y max
5
~
s
4
~
s
= =
|
|
.
|

\
|
= µ = µ
According to given results it could be concluded that process quality effectiveness of purchasing process can
be described as very high (for this specific illustrative example).

6. CONCLUSION

Improvement of key business processes is one of requests of standard ISO 9001:2008 has a critical effect on
the competitive advantage of any organization. Key business processes, their objectives, key performance
outcomes and key performance indicator present very important success factors. It is important to have in
mind that KPIs should be measured in order to provide platform for analysis of specific objectives,
benchmarking and improvement. In this paper the approach for modelling of the relative importance of
purchasing key performance indicators and their values, modelling of influence of specific objective on
purchasing process quality as well as evaluation of quality of process as whole is presented.
In this paper, it was assumed that: (a) the relative importance of key performance indicator of each objective
are determined and described by linguistic expressions; the fuzzy extent approach for the synthetic extent
values of the pairwise comparison for handling fuzzy Analytic Hierarchical Process (AHP) is used to
calculate the weight vector of key performance outcomes under each objective, (b) the values of key
performance objectives are evaluated by Business Process Management which uses linguistic expressions.
The all linguistic expressions are modelled by triangular fuzzy numbers. The proposed fuzzy model is
arranged in two interference system. In the first of interference steps, the value gap of treated key business
process is calculated using the proposed fuzzy model. In the second interference step, the process quality
effectiveness is determined on the basis of the value of gap, using the fuzzy logic IF-THEN rules. It is shown
that the proposed fuzzy model is highly suitable as a decision making tool for making decisions about the
process quality effectiveness of key business processes in any organizations.
Contributions of this paper are also the following: (1) it handles uncertainty in relative importance and values
of measures of objectives using fuzzy sets, and (2) it proposes fuzzy model for evaluation of process quality
based on fuzzy logic rules.
The proposed algorithm is tested using real-life data from service organization from central Serbia. This
procedure should be useful for the systematic application of process quality evaluation for all identified key
business processes in enterprise. All the changes, as those in the number of objectives, number of KPIs or
their relative importance and fuzzy number membership functions shape can be easily incorporated into the
model. The proposed fuzzy method for determining the process quality is the first step in improving of key
processes.


REFERENCES
Berkan, R. C., Trubatch, S.L. (1997). Fuzzy variables design in Fuzzy Systems Design Principles: Building
Fuzzy IF-THEN Rules Bases. Wiley-IEEE Press, NJ., USA.
Bhagwat R. , Sharma K. M. (2007). Performance measurement of supply chain management: A balanced
scorecard approach, Computers & Industrial Engineering 53 (2007) 43–62.
Chang, D.Y. (1996). Applications of the extent analysis method on fuzzy AHP. European J. of Operational
Research, 95 649-655.
Crosby, P.B. (1984). Quality Without Tears. McGraw-Hill, New York.
Deming, W.E. (1993). Out of the Crisis. MIT, Cambridge, Mass, USA.
Dubois, D., Prade, H. (1979). Decision-making under Fuzziness. In Advances in Fuzzy Set Theory and
Applications (ed. R.R. Yager), Ed.-North-Holland, 279-302.
Erol I., Ferrell Jr. G. W. (2009). Integrated approach for reorganizing purchasing: Theory and a case analysis
on a Turkish company. Computers & Industrial Engineering. 56(2009) 1192-1204.

Feng S., 2004. Method of Fuzzy Integrated Estimate for the Effectiveness of Quality Management System.
Manufacturing Technology & Machine Tool, 2004(2), 37-52.
Flegel M., Brozova H., 2011. Fuzzy Decision-Making for Implementing ISO 9001 and/or ISO 14001. 12th
WSEAS International Conference on Mathematics and computers in business and economics (MCBE '11),
April 11-13, Romania, 33-38.
Fore, S., 2011. Identifying quality improvement opportunities in a manufacturing enterprise. 2011 IEEE
International Conference on Industrial Engineering and Engineering Management (IEEM), 6-9 Dec. 2011,
1354 – 1358.
Holschbach E., Hofmann E., 2011. Exploring quality management for business services from a buyer's
perspective using multiple case study evidence. International Journal of Operations & Production
Management, 31(6), 648 – 685.
Johnson L. A., Chia-Yen L., 2012. Two-dimensional efficiency decomposition to measure the demand effect
in productivity analysis. European Journal of Operational Research, 216(3), 584-593.
Juran, J.M. (ed.), Quality Control Handbook, McGraw-Hill, New York, 1988.
Kaplan, S.R., Norton, P.D. (2008). The execution premium: linking strategy to operations for competitive
advantages. Harvard Business School Publishing Corporation, Boston, USA.
Khir H., Kai C., 2012. An integrated modeling method for assessment of quality systems applied to aerospace
manufacturing supply chains. Journal of Intelligent Manufacturing, 23(4), 1365-1378.
Klir, G.J., Folger, T. (1988). Fuzzy Sets, Uncertainty, and Information. Prentice Hall, Upper Saddle River,
NJ., USA.
Klir, G.Yuan, B. (1995). Fuzzy sets and fuzzy logic, theory and applications. Prentice Hall. New Jersey.
Koc T., 2007. The impact of ISO 9000 quality management systems on manufacturing. Journal of Materials
Processing Technology, 186, (1–3), 207–213.
Large O. L., König T., 2009. A gap model of purchasing's internal service quality: Concept, case study and
internal survey. Journal of Purchasing and Supply Management, 15(1), 24-32.
Lee D.-E., Lim T-K., Arditi, D., 2011. An Expert System for Auditing Quality Management Systems in
Construction. Computer-Aided Civil and Infrastructure Engineering, 26(8), 612–631.
Liu K., Pang Yanjun, Wang J., Sui. H., 2008. A New Algorithm of Fuzzy Evaluation on the Effectiveness of
Quality Management System. Manufacturing Technology & Machine Tool, 2008(12), 1-14.
Lohrmann M., Manfred R., 2013. Understanding Business Process Quality. Business Process Management,
Studies in Computational Intelligence, 444, 41-73.
Lootsma, F.A. (1997). Fuzzy Logic for Planning and Decision making, Kluwer Academic, Boston, USA.
Mahmoud B. H., Ketata R., Romdhane B., T., Ahmed B. S., 2011. A multiobjective-optimization approach
for a piloted quality-management system: A comparison of two approaches for a case study. Computers in
Industry 62(4), 460-466.
Neely, A.D., Gregory, M., Platts, K. (1995). Performance measurement system design: a literature review and
research agenda. International J. of Operations&Production Management 15 (4), 80-116.
Oakland, S.J. (2004). Oakland on Quality Management. ELSEVIER Butterworth Heinemann, UK.
Parmenter D., 2010. Key Performance Indicators (KPI): Developing, Implementing, and Using Winning
KPIs, 2 ed. Wiley.
Paulraj, A., Chen, J.I., Flynnb J., 2006. Levels of strategic purchasing: Impact on supply integration and
performance. Journal of Purchasing and Supply Management, 12(3), 107-122.
Prajogo D, Huo B., Han Y., 2012. The effects of different aspects of ISO 9000 implementation on key supply
chain management practices and operational performance. Supply Chain Management: An International
Journal, 17(3), 306 – 322.
Psomas L. E., Kafetzopoulos P. D., Fotopoulos Christos V. C., 2013. Developing and validating a
measurement instrument of ISO 9001 effectiveness in food manufacturing SMEs. Journal of Manufacturing
Technology Management, 24(1), 52 – 77.
Rezaie K., Nazari-Shirkouhi S., Miri-Nargesi S., 2011. Quality management systems and expert system
approach for designing and implementing a mechanised self-assessment system. International Journal of
Productivity and Quality Management, 8(2), 205-224.
Saaty, T.L., (1990). How to make a decision: The Analytic Hierarchy Process. European J. of Operational
Research, 48, 9-26.
Sánchez-Rodríguez C., Martínez-Lorente A, Clavel, G. J., 2003. Benchmarking in the purchasing function
and its impact on purchasing and business performance. Benchmarking: An International Journal, 10(5), 457
– 471.
Sánchez-Rodríguez C., Martínez-Lorente R. A., 2004. Quality management practices in the purchasing
function: An empirical study. International Journal of Operations & Production Management, 24 (7), 666 –
687.
Soroush A. D. , Afsaneh N. H., Napsiah I., Zulkifli L., 2012. Improving Purchasing Performance by
Implementation of QMS Process Management Approach in a Manufacturing Company, Advanced
Materials Research, vol 622-623, pp 1868-1872.
Spiegel, T., Caulliraux, M. H., 2012. Developing an Organizational Reference Model by Selecting and
Integrating Multiple References. Journal of Management Research, 4(1), 1-23.
Shih, H.S., Shyur, H.J., Lee, E.S., (2007). An Extension of TOPSIS for Group Decision Making.
Mathematical and Computer Modeling, 45(7-8), 801–813.
Tsai W-H., Choua W.-C., 2009. Selecting management systems for sustainable development in SMEs: A
novel hybrid model based on DEMATEL, ANP, and ZOGP. Expert Systems with Applications, 36(2), Part 1,
1444–1458.
Wu, J., Wang, M.Y., 2011. Improve the Quality Level of Purchasing Based on Six Sigma. Logistics Sci-
Tech, 2, 015.
Zadeh, L.A. (1975). The concept of a linguistic variable and its application to approximate reasoning,
Information Sciences, 8(3), 199–249.
Zadeh, L.A. (1977). The Concept of a Liguistic Variable and its Application to Approximate reasoning.
Information Scinence, 8 (3), 199-249.
Zimmermann, H.J. (1978). Results of empirical studies in fuzzy set theory (ed. G.J. Klir). Applied General
Systems Research, Plenum Publishing Corporation, 303-311.
Zimmermann, H.J. (2001). Fuzzy set Theory and its applications. Kluwer Nijhoff Publising: Boston.

Sign up to vote on this title
UsefulNot useful