Table of Contents
Building A System Dynamics Model
for Strategic Knowledge Management in IT Company
Sang-hyun Park*, Seung-jun Y eon**,
Dong-ho Kim***, Sang-wook Kim****
Department of MIS, Chungbuk National University
#48 Gaeshin Dong, Chsongju City, Chungbuk
Republic of Korea, 361-736
Tel : 4€2-43-261-2357
*alraview@infovil.co.kr ** naege@infovil.co.kr
**)edeo007@infovilLookr **** serra@chungbukac.kr
Abstract
TT companies make a lot of efforts for sharing and utilizing of experiences of their
members and transforming them into the organizational knowledge as a competitive
core. But they face a dilemma that they have to spend time and financial resource to
perform activities around knowedge management for the long-term gains, while
carrying field-works for making short-term profits.
As an initial attempt to tackle this managerial problem, this paper try to investigate
the mechanism of knowledge management in a small IT company in Korea with a
synthetic viewpoint using system dynamics simulation model. It depicts the dynamic
behaviors of knowedge management and presents some findings of political leverage.
Although it has to be replenished further, the scheme for the dynamism of knowedge
management and the findings presented in the paper could be useful for the decision
makers particularly of knowedge-intensive organizations
Key Words : Knowledge Management, system dynamics
1. Introduction
Many prior-researches, which emphasize the importance of knowledge management,
accentuate that it accelerates collective leaming, improves competitiveness and
facilitates responsiveness to the market changes. Unlike a financial asset that is
exhausted and devaluated as time passes, the intellectual capital provides continuous
‘value to the organization
Many companies, knowledge-intensive in patticula, recognize that the
competitiveness largely depends on their ability to create, collect and manage
organizational knowledge, and thus nm knowledge management programs for
developing, transfering, storing and disseminating the knowledge they retain. Yet,
Managing organizational knowledge as such is a challenging task, considering the
variety of factors that affect it over time, including organizational structure; informal
social processes; interactions among people, activities, and incentives, and market
changes (Hansen, Nohira & Tiemey, 1999). Moreover, it is not easy to find an optimum
level of investment of scarce resources in knowledge management for the long-term
gains, while carying fidd-worls for making short-tem pots.
To tackle these questions, this paper attempts to develop a system dynamics”
(Forester 1961; Goodman 1989; Richardson and Pugh 1981; Stenman 2000) simulation
model for the organizational knowledge management issues of a small IT company in
Korea. The paper provides a rigorous framework and the mes for qualitative
description, exploration and analysis of systems in terms of their processes, business
ules and information feedback, facilitating quantitative simulation modeling and
analysis for the design of system structure and control. Some influential factors that
accelerate or deteriorate development of the knowledge at a company were alsodeduced.
definitly with this tial.
The model in this paper covers many factors in five sectors such as personal
knowledge, organizational knowledge, projects, staffs, and finance sector that derive
from prior researches and the interviews in company and their interactions are portrayed
inameap diagramas shown in Figure 1.
© System dynamics is a methodology for understanding complex problems where there underlies
dynamic behavior affected by a certain set of feedback mechanisms. Much of the art of system
dynamics modeling (such as casual-loop diagram, stockflow diagram and simulation model) lies in
discovering and representing the feedback processes and other elements of complexity that determine
the dynamics of a system (Sterman, 2000).
(KM investment)
Personal
Knowledge
sector
KM activity time
KM activity time \
eee new staff's PK(Hire)
“experience
Productivity
Figure 1 MAP DIAGRAM OF KM
2. System Thinking approach to KM
As stated above, dynamics of knowledge management is perhaps explained by
investigating the interactiors among five sectors The authors try to make a through
review the dynamics for each sector followed by depicting it with causalloop diagram.
2.1 The dynamics of organizational knowledge
Personal knowledge contributes to the increase in organizational knowledge(Nonaka,
1991) via various KM activities. The increase in organizational knowledge in tum
increases personal knowledge over time. This reciprocal feedback structure leads to
closed loops. The behavior represented by these single loops, however, is not a good
representation of reality as anything in the real world does not continually increase over
time without balancing factors involved.
The dynamic behavior of organizational knowledge is influenced by the change of
balancing point between the acquisition and the loss. Acquisition of organizational
knowledge is accumulated in the transformation process from personal knowledge to
organizational knowledge by KM. The loss of organizational knowledge rises in
proportion to quantity of knowledge by obsolescence. Because the obsolescence of
knowledge is very fast in the IT industry which is characterized by very low inertia,
rapid technological change, and swift technological obsolescence (Pardue et al, 1999) ,
TT firms have to generate the organizational knowledge more than vanishing of that. But
as the stock of organizational knowledge grows, the time to locate relevant knowledge
also increases, and the net effect of knowledge diminishes Rich and Duchessi, 2001).
Time and monetary resources for KM activities are required to resolve the problem of
knowledge confusion (see figure 2).
Personal
Knowledge
2 OK acquisition
PK to OK, poe
7
Ag SS
An) Osganistionsl Ast) OK toss
KM investment eReee +
ee
“~
revenues
~~ * ae
(@h
KM activity time . La , _) fraction
enn
Figure 2 CLD-1: ORGANIZATIONAL KNOWLEDGE SECTOR.
The dynamic behavior of this simple system depends on which of these loops
dominates over time. If the generation of organizational knowledge is bigger than its
loss, the organizational knowledge rises (R1 and B1). Therefore the productivity from
the organizational knowledge grows quickly, and then the time to locate knowledge may
not be important for some time (R2 and B2). But KM by spending time and monetary
resourves leads to the decrease of time on work and the increase of expenditure, which
in tum mmke the revenue to go down. As a result, the effort on KM also deoeases.
2.2 The dynamics of personal knowledge
The members of organization have various kinds of knowledge such as project
execution, management experience, originality, etc. This knowledge as a whole widely
effects on their productivity and quality of work At the same time, they leam by
performing various projects, and they elevate their technical skills and increase personal
knowledge. The growth of personal knowledge positively effects on the potential
performance of prospective projects. The factors influencing on personal knowledge
increase would be summarized as follows (see figure 3):
The first is the increase in the employee’s personal knowledge usually gained in four
channels(or routes) such as experience from the previous projects(R3, experience),
education including self-study, reading etc.(R4, education), knowledge diffusion
through networks among members(R5, PK to PK), and transformation from
organizational knowledge into personal (R6, OK to PK). The second is the inflow of
personal knowledge occured by employingnew workers form outside.
FE paren statt
NOT ew |
Eatiavesiaeas compte secint——_producivdy
¥ —
~ evened
Figure3 CLD-2: PERSONAL KNOWLEDGE SECTOR.
Once personal knowledge is reserved in tacit forms, it remains usable in organization
as other organization members are able to access to and obtain it. And thus, the level of
personal knowledge would not decrease even if knowledge generator leaves the
organization. Meanwhile, secessions among the members result in diminishing the
personal knowledge, which will in tum decrease in the organizational knowledge. As
2.3 The dynamics of projects
KM policies affects on the change in both stock of knowledge available to the
organization and transformation rate of knowledge from organization to persons and
from persons to persons. The knowledge increase by KM policies increases productivity
on projects, which in tum increases the fim/s reputation in the market to get more
projects available. From the perspective of systems thinking, the increase in projects
directly leads to the increase of staffs, which in tum increase projects completed. As
projects are completed, they increase revenue, which increase investment in KM.
However, resourve allotment, especially time allocation of staffs onto the KM
activities also decreases the effort on projects that make revenue. In tum, the number of
completed projects decreases and thus revenue decreases in short term, which in result,
leads to the decrease in KM activities, and so on. This cycle indicates that the overall
effects of KM depend on how much time of staff’s to allocate onto projects and KM.
aciivities respectivedy.
revenues
gealeevenues ——_ + ee
ip
revenues gap staff demand
suai eam re, fora . ,
+
RF
‘i Gf complete project
new project project z staff
. “) Y
complete
fraction
market reputation Aer) , 7
market condition # oe
+
quality of work
% \ . er | *
productivity
Ss _| PK
Figure 4CLD-3: PROJECT SECTOR
2.4 The dynamics of staff
The staff is one of most important source of generating and transforming knowledge
in firms. Particularly, when the knowledge resourves of the firm are largely tacit, as in
the case of IT firms, tumover ratio of employees is critical in that it affects the amount
of know-how and know-who available in the fim.
When staffs leave a firm, they take their personal knowledge with them, including
that gained from work experience while at the firm (see Figure 3). To extent that the
firm relied on them as a source of knowledge, personal knowledge decreases and in tum
organizational knowledge diminishes. Their departure also gives negative impact on the
fins interpersonal network, dininating the links they provided to others in the fim.
As the departure of skilled staff directly means a loss of knowledge, it is a very
important concem of managers in IT finns. In IT industry, the annual tumover rate is
estimated to range between 10% and 20% on the average.
Some of the employees newly hired to replenish the departing staff may have work
experience and immediately contribute to the firm. Others are inexperienced, recent
graduates, who have some general skills but require training and experience to reach
proficiercy. In many firms, most entrants are of this latter type. They leam required
techniques and culture from the experienced staff inside the organization. This
adaptation causes the decrease in organizational knowledge temporarily at least and
times needed to recover.
The dynamics of staff clearly affect the collection and retention of knowledge. First,
the departure of experienced staff reduces tacit knowledge resources faster than that
provided by new inexperienced staff, putting downward pressure on organizational
knowledge. Second, the departing staff will have higher individual productivity than the
newly hiredstaff, which exerts pressure to divert more effort to knowledge management
activities to counter: As aresullt, the behavior seeks balanoe
hire delay time a. oe quit fraction
hire Aes) staff quit
‘a Oo ft
staff gap a
al ay time per staff
staff demand ee
fume resource
foo
KM activity time time ratio
Fee working _ qe
Figure 5 CLD-4: STAFF SECTOR
2.5 The dynamics of financial performance
The ultimate goal of KM is to improve the financial perfonmance of the firm. For that
reason, it is required to make a decision on the best policy to increase the financial
performance by KM. For the purpose, this paper attempts to make a solution through
system dynamics modding.
income per project net propit overhead
Retained S +
revenues
expenses
earning
=f ‘ a staff
complete project a
4e
KM investment salary 4
+ +
KM investment ratio salary per staff
Figure 6 CLD-5: PROJECT SECTOR.
A ciitical sourve of financial pressure for IT finns mainly comes from the unbalance
between the volume of project backlog and the number of workers to perform. Not only
pending projects but also new projects with a mixture of short-term and longterm
engagements are necessary to meet revenue targets and maintain a certain level of cash
flow. However, wrong staffing practices are apt to cause financial problems. Hiring staff
too slowly limits the growth of the fin; hiring staff too quickly diminishes productivity.
In response to the organizational growth, a hypothetical firm increases its staff with
inexperienced employees. These employees are not as productive as those who leave,
and are less effective in reducing project backlog. And there also are delays in
recognizing backlog and hiring additional staff. Besides, as staff increases the expense
also increases. As a result, rapid growth creates a "boom and bust" scenario in the worst
case.
KM activities often demand time and monetary resources away from doing projects
or field-works for short-term revenue. KM activities are in result regarded not only as
the source of revenue increase by achieving project quality with high productivity but
also as the source of expenditure through decreasing project quality by low intensity of
staffs.
3. Simulation and implications
Based on the dscussions thus far, a stockflow diagram (SFD) is developed for the
computer simulation runs. Exogenous variables are those related to the KM investment
amount and the time allocation ratio of staffs, which are identified as crucial factors for
KM. They ae set constant for the initial simulation runs. The other constant variables
except conditional variables in the model are developed through interviews with a finn
@s an exane and conditional variables are used by default value.
There were two sessions of computer simulation tried under different situations by
changing the values of policy leverages. Net profit reaches the peak at reasonable value
when investment ratio is 4.3% and time allocation ratio, 22% (see figure 7). On the
other hand, the highest level of organizational knowledge increased as the values
increased (see figure 8). The policy implication which result of simulation is suggesting
can be summarized as following. Firms decide reasonable levels of activity and
investment in KM. And they have to recognize that the behavior of organizational
knowledge is ahead that of revenue.
Figure 9 STOCK-FLOW DIAGRAM.
Figure 9 ORGANIZATIONAL KNOWLEDGE BEHAVIOR GRAPH
4. Conclusions and further research
The success of IT firms depends on their ability to manage organizational knowledge.
But two critical factors - ie., the activity and the investment in KM are the origin to
make not only revenue in long term but also expenditure in short term. This means it is
very important to understanding the dynamics of knowledge management for the best
10
policy. By introducing the system dynamics approach, this paper tries to explain the
dynamics of knowledge management in IT firms. The general scheme for dynamism of
KM in firms and the findings presented in the paper would perhaps provide some ideas
and directions for further study. However, it has to be admitted that the model is yet to
be refined and expanded in greater detail by identifying more variables and factors and
analyzing their dated data in more rational manner.
References
Eliot Rich and Peter Duchessi, "Modeling for Understanding the Dynamics of
Organizational Knowledge in Consulting Finns", Proceeding of the 34th Hawaii
International Conference on System Sciences, 2001.
G. Richardson and A. Pugh, Introduction to System Dynamics Modelling, Productivity
Press, 1981.
November-December 1991.
J. D. Sterman, 2000, Business Dynamics: Systems Thinking and Modeling for a
Complex World, bwin McGraw-Hill.
J. D. Sterman, "Modeling the formation of expectations: the history of energy demand
forecests", International J ournal of Forecasting 4, 1988.
J. W. Fonester, Industrial Dynamics, MIT Press: Cambridge, MA, 1961.
J. W. Forester, “Information sources for modeling the national economy", Journal of
the American Statistical Association, September 1980.
MR Goodmmn, Study Notes in System Dynamics, Productivity Press, 1989.
J. Harold Pardue, Thomas D. Clark Jr, and Graham W. winch, "Modeling shortand
long-term dynamics in the commetialization of technical advances in IT
producing industries", System Dynamics Review Val.15, 1999.
M. T. Hansen, N. Nohira, and T. Tiemey, "What's your strategy for managing
knowledge?", Harvad Business Review, Val.77, 1999.
11
Back to the Top
