SYSTEM DYNAMICS MODELLING OF INFORMATION AND
COMMUNICATION TECHNOLOGIES: POLICY STUDIES IN THE BANKING
INDUSTRY IN THAILAND
By
Mohammed Quaddus
E-mail: Quaddusm@gsb.curtin.edu.au
Arunee Intrapairot
E-mail: intrapairota@cbs.curtin.edu.au
The Graduate School of Business
Curtin University of Technology, GPO Box U 1987
Perth 6845, Western Australia
ABSTRACT
This study aims to develop a holistic, dynamic model for information and communication
technology (ICT) adoption and diffusion in a major bank in Thailand in order to identify the
various policy variables affecting the adoption and diffusion of ICTs. The generic conceptual
model of ICT adoption and diffusion was developed based on the system dynamics approach to
identify and tackle the problems and fulfil bank objectives. The SD simulation model reveals the
following issues. First, training is a vital factor for the success of technology diffusion. Second, a
backlog of problems arising from technology diffusion decreases relative advantages and sales
dramatically. Third, technology cannot bring as many customers as the bank expects due to the
constraint of market potential. Fourth, excessive investment in new technology can be detrimental
to the bank. Therefore, the bank has to determine the desired level of investment in new
technology. Fifth, without substantial investment, the bank cannot harvest economic gains from
investment in new technology.
1. Introduction
Thailand’s economy is currently confronted with a major crisis originating in the property and
financial sectors. While globalisation of the world economy through information and
communication technologies (ICT) undoubtedly contributed to the flood of capital investment,
which fuelled Thailand’s 8% annual growth since 1988, the flight of this capital made the
collapse (beginning in July, 1997) more severe, with 40% depreciation of the Baht and the local
SET share index falling to a third of its highs. To prevent repetition of such damaging outcomes
in the future, the ability of financial institution to adopt and diffuse ICT may be a necessary key
to financial control mechanisms.
Since 1983, the banking Industry in Thailand has introduced and implemented numerous of
information and communication technologies (ICT). The ICT investment commencing with
the highly popular ATM transaction services has led the industry to the advent of the
electronic banking period in Thailand (The Siam Commercial Bank (Pcl.), 1996). Adoption
of ICT is initiated generally through equipment purchases. When this happens, relevant
software applications are required. However, apart from such technological aspects,
technology adoption and diffusion should focus on managerial and organisational processes
and the social context, and should be adapted to local cultures, markets and circumstances
(Bhatnagar, 1994; Gozlu, 1994).
Currently, bank executives have to make crucial decisions in regard to adopting new
technologies, maximising utility and finding ways to promote widely those adopted fruitfully, as
well as mitigating the degree of seriousness of problems deriving from technologies, and
integrating those technologies to business performances. Consequently, this study aims to
develop a holistic dynamic model for information and communication technology (ICT)
adoption and diffusion, using one of the major banks in Thailand (i.e. the Siam Commercial
Bank, Pcl. or SCB) as a case study. The model is proposed to identify the various policy
variables affecting the adoption and diffusion of ICTs.
2. Current usage of information and communication technologies (ICT)
The study initially explores the current usage of information and communication technologies of
the bank. The results reveal various types of technologies currently employ to service customers
and facilitate work performance include ATMs, EFTPOS, smart cards, databases, data
warehouse, video conferencing, internet, Intranet banking, network systems, for instance.
Although these technologies contribute to many advantages, the information from the data
collection indicates critical problems confronting the bank such as rapid obsolescence of
adopted technologies, selection of inappropriate technologies, low productive usage of those
adopted, lack of capable employees, and high costs of technologies, coupled with unexpected
performance and low acceptance from staff and customers.
Since adopting and diffusing technology are currently routine practice for decision-makers
due to the rapid rate of technological evolution and intense competitive in the banking
industry. Therefore, in order to provide a guideline on how to adopt and diffuse technologies
productively, a generic conceptual model of ICT adoption and diffusion is proposed to capture
key variables, detect constraints and purpose leverage strategic policies.
3. Research Questions
The information and communication technology (ICT) model developed is based on the
qualitative and quantitative system dynamics approach (Coyle, 1996; Wolstenholme, 1994).
The model aims to detect the results of the following research questions
3.1 How does training support impact on technology diffusion?
Apart from technical features, important factors influencing on the success and failure of
technological implementation are organisational aspects such as training, top management
support, interactions during implementation, user involvement, and motivated and capable
users’ attitudes (Kwon & Zmud, 1987; Manross & Rice, 1986). Since innovation can succeed
only if end users have a full understanding of the technology, training is considered as a vital
policy to provide knowledge, reduce levels of resistance, create skilled human resources and
increase managerial potential (Madu, 1989). Generally, technology diffusion changes
positively with the level of training support. When technology is diffused, it creates learning
environments that convince more end users to attend training and more trained staff and
active staff further enhance diffusion rate (Quaddus, 1996).
Therefore, the hypotheses regarding training support are :
H1 a: Training support increases the rate of technology diffusion.
H1 b: Training support increases relative advantages.
H1 c: Training support increases sales.
3.2 How does backlog of problems impact on technology diffusion?
Solving work problems and reducing uncertainty in problem solving are two main reasons for
adopting information technology. However, whenever technology is diffused, a backlog of
unsolved problems associated with technology itself and organisational aspects of adopters is
created (Foschini, 1989; Gozlu, 1994; Rogers, 1983). For example, home banking which is in
operation since the 70s, not only brings about a good return on investment but also gives rise
to problems (e.g. abused information, business frauds, insecurity and unreliability, and
increasing demand for higher capacity of hardware and software) (Global Banking
Intelligence Corp., 1996). Therefore, if an organisation fails to solve a backlog of problems, it
may create one kind of uncertainty in the minds of adopters leading to demoting further
adoption and simultaneously promoting existing adopters to abandon technology use (Rogers,
1983; Saeed, 1990).
Thus, given a backlog of problems factor, the suggested hypotheses are:
H2 a: A backlog of problems decreases the rate of technology diffusion.
H2 b: A backlog of problem decreases relative advantages.
H2 c: A backlog of problem decreases sales.
3.3. How does market potential affect technology diffusion ?
Previous research indicates that early adoption of new IT applications leads to long-term
competitive advantages (e.g. market share and income) (Dos Santos & Peffers, 1995).
However, an organisation may hesitate to become involved in, or to postpone full
implementation of a particular technology because of an obscure actual demand or market
potential of a product deriving from technology use (Jirapinyo, 1997).
Market potential is physically reduced by sales in a period and increased by a flow coming
from new potential customers and customers who repurchase due to product obsolescence
(Maier, 1996; Milling, 1996). However, in reality, it is difficult to know market potential of a
particular product because many potential customers or users may decide to wait for it to
attain some initial success before entering the market. This delay occurs because early
adopters will see few benefits from the product until it is used prevalently. A wait-and-see
attitude of prospective customers may cause insufficient demand to launch the product
successfully (Caskey & Sellon, 1996). Additionally, in a dynamic environment, short product
life cycles, a sharp decline in prices and time to market also affect market potential (Maier,
1995).
In effect, although technology can be successfully diffused, economic gains are limited by the
market potential. Therefore, the suggested hypothesis is:
H3: Despite successful diffusion market potential inhibits sales.
3.4. To what extent does investment in new technology affect economic returns on
investment ?
Generally, a more costly technology is less likely to be adopted but once it is adopted, the
large investment may highly motivate diffusion (Cooper & Zmud, 1990). In the recent past in
Thailand, massive technological investment was not considered a serious issue due to the
economic prosperity of the country. However, currently, there are increasing concerns
regarding technological adoption and the overall gains in retum for such investment because
high investment not only brings many advantages but it also decreases profits (Takac &
Singh, 1992; The Siam Commercial Bank's Staff, 1998). Additionally, excessive emphasis on
technological aspects may persuade people to spend time and effort dealing with the
technology instead of dedicating themselves to their actual work performances.
It is important that an organisation has to determine a balance between desired investment in new
technology and economic retums from such investment. Therefore, the suggested hypothesis is:
H4: The bank gains higher economic retums on investment from controlled technology
expenditure than that from uncontrolled.
3.5 To what extent does the bank harvest economic gains from investment in new
technology ?
In general, a positive relationship between relative advantages and technological adoption has
been found (Kwon & Zmud, 1987; Rogers, 1983). However, economic gains from
technological investment cannot be obtained synchronously. First, when new technologies are
introduced, their potential may not be exploited fully because quite a few technologies are
implemented based on a trial-and-error basis (Gagnon & Toulouse, 1996). Second, during the
initial stages, advantages of technologies cannot be obtained or even precisely determined
whereas short run costs are readily available (Gerwin, 1988). Third, technological investment
requires an adaptation and learning process to combine environment, organisation, team, task
and technology. Once the misalignments of these factors are corrected and end users
eventually adopted, economic retums then will turn out fruitfully (Applegate, 1992).
Consequently, technologies have to be substantially invested together with minimum sufficient
usage then advantages from the technology can be harvested. The hypothesis relating to this
research question is:
H5: Economic gains can be obtained after a new technology has been substantially
invested.
4. Information and Communication Technology (ICT) Model
The information and communication technology (ICT) model is developed in two stages using
qualitative and quantitative system dynamics approach for the purpose of acquiring responses to
the identified research questions.
4.1 ICT model based on qualitative system dynamics approach
Based on extensive literature reviews and interviewing the bank staff, the model divides
organisation boundaries into four sub-sectors: the bank (i.e. technology group), bank staff,
customers, and vendors. It emphasises four resources (i.e. technology, profits, staff, and
customers), and captures main feedback loops, positive and negative, of the system (see Figure
1).
Figure 1. The influence diagram of the information and
communication technology (ICT) model
The SCB
*D=Detay
Vendors
|»
iy Ado Cr
Staff
Required Skilled +
‘Stat
Customers
According to Figure 1, the details of each feedback loop are as follows:
Negative Feedback loop A: Requirements in information and communication technologies
(ICT) of the bank are inspired by the gap between available valid ICT on the market and a
level of investment in new technology. The bank narrows its technical gap by increasing
technological investment.
Positive feedback loop B1 and B2: Investment in new technology activates the bank to
diffuse the technology in order to maximise relative advantages and customer satisfaction.
Customer satisfaction is the vital factor for generating active customers, which leads to
accelerating sales rates and profits. Ultimately, profits attract the bank to technological
expansion.
Negative feedback loop C: Massive expenditure (e.g. costs of technology and operational
costs) is accommodated throughout the processes of adopting and diffusing new technology.
Certainly, the costs reduce prospective profits.
Negative feedback loop E and F: New technology usage requires an increase in both the
number of skilled staff and levels of skill of actual staff. Failing to upgrade levels of skilled
staff widens the gap between the actual skilled staff and those required. Therefore, providing
training is necessary to fill the gap.
Negative feedback loop G: Fulfilling quality and quantity of skilled staff via training results
in increasing costs and subsequently decreasing profits.
Negative feedback loop H1 and H2: Once technology is integrated in work performances
and services, a backlog of problems begins to accumulate. If end users or customers are
annoyed or disappointed, they will abandon that technology use, Additionally, the bank may
also hesitate to adopt additional technologies.
Negative feedback loop [1 and 12: A backlog of problems exerts negative impacts on both
relative advantages and customer satisfaction. These impacts may completely or partially offset
the positive gains from previous feedback loops (Loop B1 and B2).
Negative feedback loop J: Apart from costs of technology and operating costs, disparate
costs (e.g. training costs, maintenance costs and costs from backlog of problems) continue to
accumulate with the diffusing process.
Negative feedback loop K: Although the bank deploys technology to increase customers, the
numbers of customer cannot be increased beyond market potential.
Negative feedback loop L: The bank has to find the balance between technological
investment and profits in order to arrive at the desirable level of technological expenditure
because excessive investment can reduce the bank’s profits.
4.2 ICT model based on quantitative system dynamics approach
The qualitative conceptual model identified by the above feedback loops is then quantified and
simulated using the /think software (Richmond, Peterson, & Charyk, 1994). Each feedback is
added to the simulation until the whole system is complete. This incremental technique enhances
understanding in terms of the impacts of each feedback loop, detecting errors and tracing the
logical concepts.
Initially, the feedback loops are simulated incrementally from loop A to loop C in order to
identify the behaviours of significant variables (e.g. rate of technology diffusion, relative
advantages and sales). The simulation results up to loop C are considered as baseline results and
are used to compare with other results deriving from subsequent incremental simulation.
5. Results
The simulation model reveals the following issues.
5.1 Research question1: Loop E, F and G are added to the baseline simulation for the purpose
of capturing the impacts of training support which is set up to bridge the gap of insufficient
technical skill. The derived results are then compared with those of the baseline to answer the
research question 1, How does training support impact on technology diffusion?
Figure 2: Impacts of training support on technology diffusion
a: Rate of technology diffusion
Be
: —
1- =a
According to Figure 2a, the rate of technology diffusion supported by training is higher than
the baseline for the whole periods of time. Additionally, despite an increase in training costs,
the bank still gains more relative advantages and sales (Figure 2b and 2c.).
Therefore, it can be concluded that training is a vital factor for technology diffusion and all the
three hypotheses are supported. That is, training support increases the rate of technology
diffusion, relative advantages and sales.
5.2 Research question 2: As can be seen in Figure 1, a backlog of unsolved problems
directly affects rates of technology adoption and diffusion, consumes costs for resolving
problems and possibly reverses relative advantages and customer satisfaction from positive to
negative outcomes. The entire feedback loops (from Loop A to Loop J) are then simulated to
detect the impacts of the backlog of problems.
According to Figure 3a, a backlog of problems, without training support, decreases the rate of
technology diffusion. However, the result is not significantly different from that of the
baseline because the percentage of a backlog of problems that the bank has currently
confronted is low (The Siam Commercial Bank's Staff, 1998). Figures 3b and c reveal that
relative advantages and sales do decrease because of the impacts of a backlog of problems.
Figure 3: Impacts of a backlog of problems on technology diffusion
2a: Rate of technology diffusion
be: Relative Advantages eles
In effect the end results reveal that a backlog of problems exerts impacts on technology
diffusion. All three hypotheses are supported; that is, a backlog of problems decreases a rate of
technology diffusion, relative advantages and sales. It is interesting to observe that, apart from
sales, the rate of technology diffusion and relative advantages, given the training factor, are still
higher than those from the baseline because of the more powerful influences of training support.
5.3 Research question 3: A feedback loop K is added in order to observe the influence of market
potential on sales. As can be seen from Figure 4a and b, although the rate of technology diffusion
is similar to those of previous simulations, sales increase until the market potential is completely
absorbed.
Therefore, it can be concluded that technology cannot increase sales beyond market potential,
although it can be successfully diffused.
Figure 4: Impacts of market potential on sales
2: Rate of techncogy fusion bSalos
5.4 Research question 4: Investment in new technology to eliminate technological gap or
catch up with technological evolution cannot be infinitely increased because technology
expenditure relates directly to profits of an organisation.
Loop L is added to the simulation to observe the impacts of the bank’s control of
technological investment based on the amount of its profits.
Figure 5: Impacts on profits of controlled investment in new technology
8000
7000
6000
5000
4000
[uncontrolled Tech Expense
LE = Controlled Tech Expense
3000
2000
1000
500 1000 1500 2000 2500 3000 3500 4000 4500 5000
Investment
According to Figure 5, with the same amount of technological investment, profits deriving
from controlled technology expenditure are higher than those without control. Hypothesis H4
is therefore supported.
5.5. Research question 5: The simulation results of comparing investment in new
technology with profits in figure 5 also reveal that, the bank cannot gain any profits from its
technological investment at the initial period. Profits increase after the bank has spent
substantial money for technology investment. Hypothesis H5 is therefore supported.
6. Conclusions
The ICT model informs the following issues. First, training support has the potential to accelerate
the rate of technology diffusion and economic gains whereas a backlog of problems hinders them.
Second, market potential constrains an increase in economic retums although technology is
successfully diffused. Third, it is important to determine the balance between the desired
investment in new technology and its prospective outcomes because massive investment in new
technology does not always bring a good retum on that investment. Fourth, economic gains from
new technology are obtained after an organisation has spent substantial resources on technology
investment.
This model enables bank officials to understand the present state and constraints of technology
adoption and diffusion. This can be further used for policy analysis and forward planning to
mitigate the constraints and re-design the system behaviours. Since existing and potential
banking technologies are abundant, the model can be initially used to gain holistic understanding
before applying any particular technologies or tailoring for specific organisations.
Apart from interviewing data, this proposed conceptual model attempts to include numerous
variables based on literature review. However, in reality, relatively few variables are taken into
account by bank staff. Additionally, complete data are hard to obtain due to their physical
properties (e.g. implicit, intangible and unclassified). Therefore, for future research, this
conceptual model will be elaborated into two specific technologies (i.e. Intranet banking and data
warehousing technology) using only variables that the bank has considered in order to improve
the model to fit with reality.
References
Applegate, L. D. (1992). 4 case study in the assimilation of technology support for teams.
New Y ork: Van Nostrand Reinhold.
Bhatnagar, S. C. (1994). Is technology transfer the answer? Paper presented at the IFIP, 13 th.
World Computer Congress 94, Amsterdam.
Caskey, J. P., & Sellon, G. H. (1996). Is the debit card revolution finally here? In P. O'Hara (Ed.),
A book of readings on the financial system (6 ed., pp. 79-95). Perth: Curtin University of
Technology.
Cooper, R. B., & Zmud, R. W. (1990). Information technology implementation research: A
technological diffusion approach. Management Science, 36(2), 123-139.
Coyle, R. G. (1996). System dynamics modelling: A practical approach. London: Chapman &
Hall.
Dos Santos, B. L., & Peffers, K. (1995). Rewards to investors in innovative information
technology applications: First movers and early followers in ATMs. Organisation
Science, 6(3), 241-259.
10
Foschini, S. (1989). Development and use of system dynamics models as tools for strategic
planning of flexible assembly systems. In P. M. Milling & E. O. K. Zahn (Eds.),
Computer-based management of complex systems (pp. 89-96). Berlin: Springer-V erlag.
Gagnon, Y. C., & Toulouse, J. M. (1996). The behaviour of business managers when
adopting new technologies. Technological Forecasting and Social Change, 52, 59-74.
Gerwin, D. (1988). A theory of innovation processes for computer-aided manufacturing
technology. IEEE Transactions on Engineering Management, 35(2), 90-100.
Global Banking Intelligence Corp. (1996). Technology Investment in Banking
http://www.globalbanking.com/tech.htm
Gozlu, S. (1994). Transfer of information technology to a developing environment: The Turkish
case. Paper presented at the IFIP, 13 th World Computer Congress 94, Amsterdam.
Jirapinyo, P. (1997). The SCB banking technologies . Personal Communication, May 19, 1997
Kwon, T. H., & Zmud, R. W. (1987). Unifying the fragmented models of information system
implementation. New Y ork: John Wiley.
Madu, C. N. (1989). Transferring technology to developing countries: Critical factors for success.
Long Range Planning, 22(4), 115-124.
Maier, F. H. (1995). Innovation diffusion models for decision support in strategic management.
Paper presented at the 1995 Intemational System Dynamics Conference, Tokyo, Japan.
Maier, F. H. (1996). Substitution among successive product generations: An almost neglected
problem in innovation diffusion models .
Manross, G. G., & Rice, R. E. (1986). Don't hang up: Organisational diffusion of the intelligent
telephone. Information & Management, 10, 161-175.
Milling, P. M. (1996). Modelling innovation processes for decision support and management
simulation. System Dynamics Review, 12(3), 211-234.
Quaddus, M. A. (1996). GSS supported system dynamics: A model for IT planning.. Perth,
Westem Australia: Curtin University of Technology.
Richmond, B., Peterson, S., & Charyk, C. (1994). Introduction to system thinking and ithink.
Hanover, NH: High Performance Systems.
Rogers, E. M. (1983). Diffusion of innovations. (3 ed.). New Y ork: The Free Press.
Saeed, K. (1990). Managing technology for development a systems perspective. Socio Economic
Planning Science, 24(3), 217-228.
Takac, P. F., & Singh, C. P. (1992). Banking technology: Improving its potential through better
management. Management Decision, 30(5), 17-20.
The Siam Commercial Bank (Pcl.). (1996). http://www.scb.co.th.
The Siam Commercial Bank's Staff. (1998). Personal Communication. January 1998.
Wolstenholme, E. F. (1994). System enquiry: A system dynamics approach. Chichester: John
Wiley & Sons.
i
