A System Dynamics Approach for Knowledge Management
and Business Excellence: An Application in Iran
Mohammad Dehghani Saryazdi’, Mohammad Saleh Owlia” and
Jamal Hosseini Azabadi**
'P.H.D. candidate in Industrial Engineering, Department of Industrial
Engineering, Yazd University, Yazd, Iran
Tel: +98-913-1594171
E-mail: ie.dehghani @ gmail.com
? Associate Professor, Department of Industrial Engineering, Yazd
University, Yazd, Iran
Tel: +98-351-8122416
E-mail: owliams @ yazduni.ac.ir
°M.Sc. in Industrial Engineering, Department of Industrial Engineering,
Iran University of Science and Technology (UST), Tehran, Iran
Tel: +98-913-1594808
E-mail: hosseini.j @ gmail.com
Abstract
One of the excellence enablers is KM'. In order to evaluate the KM processes, a
comprehensive model is required, which should be able to capture all aspects of KM.
One of such models is KMAT?. This research exploits system dynamics in order to
measure the effects of KM on business excellence with a combination of KMAT and
EFQM? Relationships between KM and EFQM are analyzed and demonstrated by
means of the literature reviews, expert interviews and system dynamics .The results of
this study could be useful for knowledge management planners and managers in
organizations.
Key words: Knowledge Management; Business Excellence; System Dynamics
“Corresponding Author
' Knowledge Management
? knowledge Management Assessment Tool
3 European Foundation for Quality Management
1. Introduction
In a world of dynamic and discontinuous change, organizations are constantly seeking
ways to adapt themselves to new conditions so that they are prepared to survive and
flourish in a competitive marketplace (Albert 1997). The proliferation of the knowledge
economy (Castells 1996), emphasizing the value of information as an enabler of
competitive advantage, is naturally driving many companies to re-examine the ways
they have treated their knowledge assets in the past and to identify ways in which they
can exploit them more effectively in the future.
Drucker(1993) has described knowledge, rather than capital or labor as the only
meaningful economic resource in the knowledge society, and Senge(1990) has warned
that many organizations are unable to function as knowledge based organizations,
because they suffer from learning disabilities those organizations that will succeed in
the global information society are those which can identify, value, create and evolve
their knowledge assets (Rowley 1999; Lai 2007).In such a landscape, it is not surprising
that knowledge and information management has emerged as one of the most popular
strategic change management approaches in the dawn of the twenty-first century
(Davenport and Prusak 1997). Its supporters argue that organizations may achieve
significant competitive advantages by analyzing the data and information that often
remain unexploited in organizational systems and by transforming them into useful and
actionable knowledge (Giaglis 2002). Although there could be found some researches
about the improvement and application of EFQM model and KM (Jager 1999; Yim et
al. 2004; Bose 2004; Kumar 2010; Pal et al. 2004; Smits and Moor 2004; Waraporn et
al. 2010) and also about utilization of system dynamics in EFQM model(Dehghani et
al. 2009; Eskildsen et al. 2001; Sadeh and Arumugam 2010) in the literature, there have
been no attention paid to simultaneous investigation of EFQM and KM for finding the
effects of KM on business excellence. Accordingly, this research tries to depict these
effects by combining the EFQM model and KM Assessment Tool. The conducted
studies that exploit system dynamics in the knowledge management are as follows:
e Drew and A. Smith (1996) in which, with regard to the importance of knowledge
resources (as a competitive advantage) in increasing the market share of a
business, it is stated that the nature of these intellectual capitals and the
interactions of their system dynamics are recognized weakly yet.
e Eklof et al. (2004), the main goal of which is to survey the knowledge
management in a Law Firm with focus on the supportive role of information
technology. For this aim, system dynamics simulation tool is applied to present
diagrams of cause and effect loops, stock and flows, in order to describe
different variables and their effects on each other. These diagrams indicate
variables influencing the general level of organization’s knowledge and the need
to knowledge management.
Knowledge could be managed regarding different aspects of process, leadership,
culture, technology, and measurement (i.e. KMAT dimensions) and its effects on results
components of EFQM, including customers, employees, society, and key performance
indicators, could be assessed. The proposed model could not illustrate the effectiveness
of KM unless it is applied and evaluated regarding apt performance indicators. These
N
indexes should be of both types of quantitative and qualitative and should be measure
via a systematic approach. This is the only way managers can understand the critical
success factors of KM in their organizations. One of the appropriate systematic
approaches is system dynamics, which provide the opportunity to model and analyze the
behavioral patterns in phenomena with the goal of decision making based on prediction
(Jager 1999; Yim et al. 2004).
This research aims at developing a dynamic model for measuring the effectiveness of
Knowledge Management (KM) processes on excellence of organizations. It focuses on
analyzing the key aspects of KM which affect business excellence. This analysis would
be conducted through simulation and by using the mentioned dynamic model. The case
study will describe in this article is concerned about the Pars Refractories company in
Iran. For relating the KM and the business excellence, it is required to understand and
establish the “cause and effect” relationships between the KMAT dimensions and
results indexes of EFQM. Systematic approach shed light on the cause and effect
relationships. Furthermore, it states that all diverse aspects and sections of an
organization are related to each other and nobody could improve on section or whole of
an organization without enhancing others. On the other hand, amongst the numerous
variables and their relations, only some especial cause and effect loops dominate and are
significant to overall behavior of the system. In this paper, after defining the
relationships between variables and formulating a combinational dynamic model for
measuring the effectiveness of KM, different policies are designed for developing KM
plans and their results are evaluated. Amongst the various methods for modeling system
dynamics, a simple one consisting of problem definition, cause and effect diagram
modeling, dynamical model generation, simulation, analysis, and application steps is
utilized.
2. Theoretical Bases
2.1. The EFQM Excellence Model
European Foundation for Quality Management (EFQM) has nine criteria. Five criteria
are known as enablers and four other ones are called results. "Enablers" cover what an
organization performs and "results" include what an organization obtains. "Results" are
obtained by implementing "enablers" and "enablers" are improved by getting a feedback
from "results" (Dehghani Saryazdi 2006).Enablers include leadership, strategy, people,
Partnerships & Resources and Processes, Products & Services. Results consist of the
results for customers, people, society and key results. All enablers, except strategy,
include 5 sub-criteria. Strategy criterion includes 4 sub-criteria. Each of results also
includes two sub-criteria. Thus, 24 sub-criteria and 8 sub-criteria respectively have been
defined for enablers and results.
2.2. KMAT Model
The knowledge management a:
make an initial high-level as:
ssment tool (KMAT) is designed to help organizations
ment of how well they manage knowledge. Completing
the KMAT can direct organizations toward areas that require more attention, as well as
identify knowledge management practices in which they excel.
2.3. System Dynamics
System dynamics is an approach to understanding the behavior of complex systems over
time. It deals with internal feedback loops and time delays that affect the behavior of the
entire system (Sterman 2000).
3. Modeling Process
3.1 Cause-and-Effect Diagram Modeling
In the first phase, the cause-and-effect diagram of the combinational model of KMAT
and EFQM is designed regarding the identified variables (see figure 1). It can be seen in
this diagram that in the forward direction the business strategy and policy affects the
KMAT dimensions and consequently KMAT dimensions influence the performance
indicators of the results section of EFQM, which in its turn changes the outcome
indexes of the results section. Furthermore, in the backward direction, the performance
and outcome indexes influence KM separately and KM impacts the business strategy
and policy. Thus, these forward and backward paths create some loops in the model.
Naturally, if the business strategy and policy are improved, knowledge would be
managed better and when the KM in enhanced, the performance indicators would be
amended in the result section. Increasing the values of performance indicators, results in
the raise of outcome indexes. On the other hand, the better the KM, the more
appropriate the business strategies and policies, and superior results decrease the need to
KM improvement. Accordingly, the focus should be on the processes for which the
results (i.e. performance indicators and outcome indexes) are smaller. So, the loops of
7
this diagram are negative (balancing) feedback loops which are illustrated byCh
It is notable that changes occurred for KMAT dimensions impact the performance
indicators after a delay, like what happens between performance indicators and outcome
indexes in the results section of the EFQM model. Other relationships have specific
delay times too. In the diagram these delays in relationships between two variables are
shown by two parallel lines (ll) on the corresponding arrows. Furthermore, in the
designed model three extra variables namely, customer loyalty, image, and employee
satisfaction are included which will be elaborated in the next section. Regarding the
relationships between variables the following points could be seen in figure 1:
e Because one of the dimensions of KMAT is measurement, so this variable is
included in the performance and outcome indicators of EFQM model and there
is no obligation for defining a new variable under the name of ‘measurement’.
e Regarding that the business strategy and policy affects the leadership dimension
of KM, and because leadership, on its own turn, affects the dimensions of
processes, technology, and organizational culture, their impacts are illustrated in
the cause-and-effect diagram through the relationships between them. On the
other hand processes, technology, and organizational culture influence the
results indexes and increasing the amounts of result indexes decreases the
required effort for KM.
Performance indicators affect the outcome indexes. Customer loyalty is
influenced by outcome indexes of customer results. Image is affected by
outcome indexes of customer and society results. Employee results impacts
employee satisfaction. The culture variable influences the employee results
index. Technology impacts the society results indexes and key performance
indicators. Customer results, society results, and key performance indicators are
impacted by processes variable.
In the model, letter ‘a’ indicates the outcome indexes and letter ‘b’ addresses
performance indicators.
In the model, processes, leadership, culture, and technology variables are
exhibited by signs I, II, III, and IV respectively.
In the proposed model, the measurement variable includes all the indexes of the
EFQM results section (i.e. customer results, employees, society and key
performance indicators).
eae: adjust Key Perform
é lance Result (a)
\ Key Performance
Rese (a)
Figure 1. Cause-and-Effect diagram
3.2 Stock and Flow Map Modeling
In this
section the quantitative relationships between model variables are defined. Here,
the time period is set to one year. The model is executed with Vensim PLE software
application. Some of the relationships of the dynamical model are included as appendix.
The model is simulated for 10 years beginning from year 2010. Since KMAT and
EFQM models evaluate the organization in a specific point of time, all the sub-criteria
of the model are of auxiliary variable type.
For the cumulative trend of organizational progress to be evaluated, it is required to
define levels in order to show the trend of organizational progress through the time.
Based on this it is possible to define some levels regarding all the results indexes which
show the progressive business trend during the time window. For increasing the model
performance, in this paper level of image, level of customer satisfaction, and level of
employee satisfaction, which are achievable from the resultant of model indexes, are
defined. These levels help the organization significantly by showing the excellence
path, because the make it possible to see the cumulated effects of organization’s
strategies, systems, and activities through the time, which illustrate the past and current
organizational performance. In each time period the image level is resulted from
summation of image rate, the inputs of which are outcome indexes of Customer Result
(a) and Society Result (a). Moreover, in each time period the level of customer loyalty
is the result of summing customer satisfaction rate which has outcome indexes of
Customer Result (a) as its inputs. Furthermore, level of employees’ satisfaction is
achieved in each period by summing up employees’ satisfaction rate inputs of which are
outcome indexes of Human Resource Result (a).Based on the definitions of dynamical
equations and relationships between variables, the dynamical model of figure 2
emerged.
‘Resa
Figure 2: Stock and Flow map
4. Performance Tests of the Developed Model
In this research, diverse types of tests such as units’ consistency test, collaborative error
test, scope sufficiency test, parameter evaluation test, structure evaluation test, and
boundary conditions test are used in order to evaluate the model performance. These
tests are described below:
e Unit consistency test: Our model passed this test while all of its units were
approved by Vensim PLE software when the Units Check option was active.
¢ Collaborative error test: the proposed model is independency to time unit. For
example, if the time unit is assumed “one year” initially, if it is changed to
“Quarter” the model should generate quite similar results. The results indicated
no changes in the behavior of the mentioned variable in different time units as
illustrated in figure 3.
Customer Result(b)
80
0
2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020
Tame (Year)
"Customer Resui(b)" = Quarter
"Customer Resul(b)" = one year
Figure 3. Behavior of Customer Results (b) in different time units
e Scope sufficiency test: This test was passed through further surveying the EFQM
and KMAT models, which involve defined criteria, sub-criteria, and indexes.
e Parameter evaluation test: the expert opinion is used in all the variables of the
model as an estimation of all parameters.
.
Structure evaluation test: This test was passed by the consistency of models
behavior with its structure. Because the variables of the model create negative
feedback loops, they should be goal-seeking. For example, the goal-seeking
behavior of the variable Customer Result (b) is illustrated in figure 4.
Customer Result(b)
0
o
40
2
0
2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020
Tame (Year)
‘Customer Resul(b)": KMS.
Figure 4. Goal-seeking behavior of variable Customer Results(b)
e Boundary conditions test: This test was conducted for the model and its
performance was approved in the boundary conditions. For example, the
amounts of policy, strategy, and leadership variables were tested in the boundary
values of 0 and 100 and their effect of Customer Result (b) variable was
captured. The results indicated no changes in the mentioned variable in the
boundary conditions as illustrated in figure 5.
Customer Result(b)
; Bn5>-Saene
40
o
2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020
Tine (Year)
"Customer Result)": KMS 100
‘Customer Result)": KMS 0
Figure 5. Behavior of Customer Results (b) in boundary conditions
5. Policy Making
For evaluating diverse policies by the generated dynamical model, an ‘improvement’
index is defined for variables I, II, II, IV, and Association’s Strategy & Policy. The
amount of improvement is measured for each policy regarding business goals and
through the execution of the model for that policy. At the end the best policy is
identified. Regarding the resource limitations and the variety of adaptable strategies in
the organization, the resource should be distributed amongst all the five dimensions of
KM in such a way that the maximum efficiency is achieved through out whole the
model in order to have the best trends for customer loyalty level, employee loyalty
level, and image. The chief variables selected in this research were customer loyalty
level, employee loyalty level, and image. Four policies described below were simulated:
First Policy: This Policy is called “Technology-based KM Approach” and highly
attends to technologies required for KM. Technology plays important role in the
advance of KM and its effect on result indexes yields in organizational excellence. In
this policy the significant variable is variable IV, so the “Improvement IV” variable,
indicating the amount of variable IV, is increased 50%.
Second Policy: This Policy takes the “Organizational Culture-based KM Approach” and
attends to culture making and its penetration in the organization. Employees are the
most valuable resources of an organization who should adapt the knowledge sharing
culture. In this Policy the significant variable is variable III, so the “Improvement IIT”
variable, indicating the amount of progress of variable III, is increased 50%.
Third Policy: This Policy is called “Process-based KM Approach” and highly attends to
KM processes. Proce: play important role in achieving KM goals. This Policy
emphasizes variable I, so the “Improvement I” variable, indicating the amount of
progress of variable I, is increased 50% in this Policy.
Fourth Policy: This Policy is called “Overall KM Approach” and highly attends to all
the three aspects of technology, culture, and KM processes simultaneously. Accordingly
in this Policy the significant variables are variables I, III, and IV. In this case regarding
the limited resources and the need to hold a trade-off between all the three aspects, the
organization should have a slower and equal progress in all of them. So, all the
“Improvement” variables are increased 20%.The results of the Policies are illustrated in
figures6 to 8.
LEVEL OF EMPLOYEES LOYALTY
200
150
100
50
o
2010 2011 20122013 2014 2015 2016 2017 2018 2019 2020
Tare (Year)
LEVEL OF EMPLOYEES LOYALTY :|) +++ + 4 +
LEVEL OF EMPLOYEES LOYALTY :2. 22 2 2 2
LEVEL OF EMPLOYEES LOYALTY :3 ——2——2 3 2 2 —
LEVEL OF EMPLOYEES LOYALTY :4
Figure 6. Trends of employee loyalty level for different policies
LEVEL OF CUSTOMERS LOYALTY
2010 2011 2012 2013 2014 2013 2016 2017 2018 2019 2020
Tare (Year)
LEVEL OF CUSTOMERS LOYALTY : 1) —-—+——+ +++
LEVEL OF CUSTOMERS LOYALTY :2. 2222»
LEVEL OF CUSTOMERS LOYALTY :3. —-s——s——s 2s —
LEVEL OF CUSTOMERS LOYALTY :4
Figure 7. Trends of customer loyalty level for differ level for different policies
LEVEL OF IMAGE
2010 2011 3012 2013 2014 2015 2016 2017 2018 2019 2020
‘Time (Year)
LEVEL OF MAGE:
Figure 8. Trends of organizational image for different policies
In figure 6, it could be seen that amongst the four above policies, the second one,
“Organizational Culture-based KM Approach”, illustrates the best trend. Accordingly, it
could be concluded that organizational culture, amongst the KM aspects, affects the
progress of employee loyalty more strongly. After the second policy, the fourth, the
first, and the second policies exhibit the best trends respectively.
Figure 7 shows that amongst the four above policies, the third one, “Process-based KM
Approach”, illustrates the best trend. Accordingly, it could be inferred that, amongst the
KM aspects, the KM process influences the progress of customer loyalty more strongly.
After the third policy, the fourth, the second, and the first policies exhibit the best trends
respectively.
In figure 8, the third policy or “Process-based KM Approach”, exhibits the best trend.
Thus, it could be implied that KM processes affects the progress of level of image more
than other KM aspects. After the third Policy, the fourth, the second, and the first
Policies respectively show the best trends. Finally comparing all the four Policies it
could be summed up that: if the highest priority in the organization belongs to employee
loyalty, the KM approach should be based upon organizational culture and if the
organization aims at improving customer loyalty and image level, then the KM
approach should be pivoted around process.
6. Conclusion
Organizational knowledge is very complex and has multiple dimensions. Several
scholars have emphasized the importance of internal and external knowledge and this
notion has important implication for knowledge management; the knowledge residing
outside the boundaries of organization must be managed along with the internal
knowledge. Therefore, organizations need to understand the dynamics of their
knowledge capital and knowledge acquisition policies to achieve better organizational
out comes. The model suggested in this research demonstrates the relationships between
the KM sections and results indexes. Thus, this research was an endeavor to model the
effects of KMAT dimensions on EFQM results through system dynamics modeling
approach, and finally analyze the change trends for different values of variables. The
chief benefits of this model are:
e In the generated model the time distance between the effect of cause and
appearance of its effects is mentioned by including delays.
e Applying the “What happens if”. This action reduces the risk of program failures
before implementing them. In the proposed model four Policies are designed and
examined in order to find the best one.
Areas for further research could be:
e Regarding that in this research the KM and business excellence dimensions are
considered generally, in the future researches these concepts should be
elaborated more in order to improve the relationships.
e¢ Modeling could not be of benefit for organizations solitarily, but it should be
conducted along with execution and customization.
References
Albert, S.1997.Managing Knowledge: Experts, Agencies, and Organizations,
Cambridge University Press.
Bose, R. 2004. Knowledge management metrics, Industrial Management & Data
Systems, 104(6), pp. 457-468.
Castells, M. 1996. The Rise of the Network Society, Blackwell Publishers, Cambridge,
MA.
Davenport, T.H., Prusak, L.1997. Working Knowledge: How Organizations Manage
What They Know, Harvard Business School Press, Boston, MA.
Dehghani Saryazdi M., Owlia M.S., Kiani B., Noughandarian K.2009. Analysis of
EFQM excellence model using system dynamics, Iranian Journal Of Management
Sciences Winter, 3(12), pp.65-82.
Dehghani Saryazdi, M. 2006. Performance Assessment of Yazd Regional Electric Co.
Based on EFQM excellence Model. B.S. Dissertation in Industrial Engineering,
Yazd University, Iran.
Drew Stephan A., Peter A. Smith.1996. Competitive Advantage Through Knowledge
Management: A System Daynamics Approach, Fourtheen International Conference
of the Systems Daynamics Society, Cambridge, Ma.
Drucker, P. 1993. Post Capitalist Society, HarperRow, New York, NY.
Ekl6f Frida, Hilde Fiskum, Lise Haalien, Elisabeth Haugland, Bjgrn Melling, Erik
Spieler, Anne Lene Sveren, Oleg Tukh. 2004. System Dynamics Simulation for
Knowledge Management in a Law Firm,BI Sandvika Master of Science in Business
GRA 6643 Knowledge Management Systems.
Eskildsen, Jacob K., Kai Kristensen, Hans Jorn Juhl.2001. The Causal Structure of The
EFQM Excellence Model, International Journal of Quality & Reliability
Management, 18 (8), pp.783 - 795.
Giaglis, G. 2002. “European research on knowledge and information management:
current status and future prospects, available at:
www-.eltrun.gr/papers/ICSTM_Giaglis_KIM.pdf
Jager, Martha de. 1999. The KMAT: benchmarking knowledge management, Library
Management, 20(7), pp. 367-372.
Kumar, S. A. 2010. Knowledge management and new generation of libraries
information services: A concepts, International Journal of Library and Information
Science, 1(2), pp. 024-030.
Lai, L.F. 2007. A knowledge engineering approach to knowledge management,
Information Sciences, 177 4072-94.
Pal, N., Sundaresan, s. Ray, J. Bhargava, H., Glantz, E.. McHugh
M.W.2004.Knowledge Quotient™ (KQ): A Way to Measure the Knowledge
Intensity of Your Team, The Penn State E-Business Research Center.
Rowley, J. 1999. What is knowledge management?, Library Management, 20 (8), pp.
416-19.
Sadeh, E., Arumugam, V.C. 2010. Interrelationships among EFQM Excellence Criteria
in Iranian Industrial SMEs, European Journal of Economics, Finance and
Administrative Sciences, ISSN 1450-2887, (19).
Senge, P. 1990. The Fifth Discipline: The Art and Practice of the Learning
Organization, Currency/Doubleday, New York,NY, August.
Smits, M.T., & Moor, A. de. 2004. Effective knowledge management in knowledge-
intensive organizations, Proceedings of the Sth European Conference on
Organization Knowledge, Learning and Capabilities, pp. 1-10.
Sterman, J. D. 2000. Business Dynamics: Systems Thinking and Modeling for a
Complex World, Irwin McGraw-Hill, London.
Waraporn, P., Meesad, P., and Clayton, G. 2010. Proposed ontology based knowledge
acquisition and integration framework for clinical knowledge management,
IJCSNS International Journal of Computer Science and Network Security, 10 (3).
Yim N.H., Kim S.H., Kim H.W., and Kwahk K.Y. 2004. Knowledge based decision
making on higher level strategic concerns: system dynamics approach, Expert
Systems with Applications, 27, pp. 143-158.
12
