To Main Proceedings Document
Developing a model for paradigm shift in service industry
Hazhir Rahmandad
Sharif University of Technology- Industrial engineering student
E-mail: rahmanra@ ind.sharif.ac.ir
January 1999
Abstract
For a long time service industry has been successfully following a mass production manufacturing model,
But in last decade, this strategy has lost its effectiveness, putting many successful companies under
pressure. Instead a new model has grown to be effective, taking place of old approach in managing service
companies. This paper discusses and models the process of above paradigm shift using a system dynamics
point of view.
Key words: Service industry, success chain, failure chain.
Introduction
In last half century, service industry has grown a lot, gradually becoming the largest part of US economy (
Quinn & Gagnon, 1986). This fast growth is based largely on mass production manufacturing model, in
which the main attempt of management is to keep a quick, uniform and clean service (Schlesinger &
Heskett, 1991). For example consider usual supermarkets or fast-food companies such as MacDonald's. In
this case we can easily see the trend discussed above: for a long time MacDonald’s has been a model of
efficient and successful service company not only for fast-food operators but also for hotels, retail stores
and other businesses where personal contact of personnel with customers plays a great role in the job. In
the late 1980's, the unreachable growth and profitability of this company started to stagnate and even fall,
the company had much more problem finding satisfactory employees while giving up lots of customers to
companies offering more varied menus. A pplying traditional solutions, they put more pressure to increase
their advantages in the technological field but this even made the situation worse. So what is the problem?
Finding the underlying reasons of such stories has been an important field of research in recent years.
These reasons are implying some underlying structures which would be very well discussed using SD. The
main attempt of this paper is to adopt results of former researches into SD and build a model capable of
supporting further microworlds and management games. The paper tends to give a dynamic picture of
these theories, investigate the ability of them in causing the paradigm shift’s reference mode and build a
model capturing “soft “ characteristics of service industry. This would give insight for management
purposes through steps taken to build the model. According to importance of service industry, it would be
necessary for managers of this section to understand the structures causing this paradigm shift in order to
adopt wiser and more efficient policies.
Through the rest of this paper we would take a better look at the problem and discuss theories explaining
it, then based on these theories a system dynamics model is built to analyze the dynamics of service
industry, Finally we are going to analyze the behavior of the model and discuss some policy issues.
Historical Development of the Theory
As mentioned above, service industry has developed using a strategy mainly taken from mass production.
In fact most successful companies, has been investing much on expanding technology in their job so that
they can restrict personal contact with customers as much as possible. In this way, they needed less skillful
personnel which would cost them less and they could minimize their variable costs using automated
machines and uniform processes (resulting in uniform service) (Schlesinger & Heskett 1991). For a long
time this policy was successful: our picture of big chain supermarkets, banks, fast-food stores and even
hotels indicate the same story. But gradually the profitability of service industry attracted many new
competitors with new ideas to improve the quality of service as a weapon for improving their market
share. This approach, increased the expectation of customers in a way that today old approach seems to
have lost its efficiency: most powerful service companies of last decades, are facing lots of problems. They
are losing customers while facing a big turnover. On the other hand there has evolved some new
successful companies with a new approach. They value their personnel much more, hiring experienced
men with great communication skills so that they can keep a good personnel-customer relationship. These
companies avoid replacing their personnel with machines, and try to give much more flexible service.
Their personnel have more freedom to solve customer problems as fast as possible and they have linked
compensation to performance at all levels of the company. This new strategy has been rewarded by fast
growth of these companies gaining higher and higher market shares. Figure 1 shows a typical reference
mode for the problem.
Reference mode of system behaviour
1940 2000
Time (Month)
New Companies ———e Size of Company
Old Companies —__—_ Size of Company
Figure 1
Why this new approach is more and more successful so that replacing old method?
Personal contact is a key element in service industry, in fact it is probably as important as the service you
are buying itself, so having skillful front-line workers is a great advantage. Customers prefer the service as
they like it, not in the way it is designed, and they want front-line workers to have the freedom to solve
their problems as fast as possible instead of referring them to supervisor for anything out of regular. In
this way customers served with new method are more satisfied and therefor more loyal, and it is the key to
profitability in service industry (Heskett, Jones, Loveman, Sasser & Schlesinger 1994). But what happens
to increased salary costs? Although experienced service men need more salary, they acquire much less
supervision, while being more profitable compared to rookies (they can serve more customers with better
quality). So while they are getting more salary, you can decrease your management layers ( Schlesinger &
Heskett 1991). On the other hand, most service jobs used to be dead-end ones which were filled by young
untrained guys, staying in the position not for more than a year. This means a great turnover cost which
can be cut down using professional employees and keeping them satisfied in a quality company ( Heskett,
Jones, Loveman, Sasser & Schlesinger 1994) . Studies (the same reference) show that, profitability comes
mainly from customer loyalty (compared to the number of new customers), and this is driven by customer
satisfaction of service. Satisfaction is a result of value added to the service which would come from good
communication and personal contact when serving the customer or the great number of choices the
customer have when served (imagine a restaurant with variety of menus compared to one serving a few
kind of foods). Y ou can increase value added to the service with help of productive employees, those who
serve the customer kindly trying to satisfy him in any way they can. But only satisfied, well-trained
personnel would be capable of such a behavior. But how can we have more satisfied personnel? It is a
common belief to link satisfaction of employee to his salary, but studies show that satisfaction is more
driven by factors such as level of liberty of personnel in their job and for implementing new ideas, or how
friendly and satisfactory is the work environment. We would call these factors internal quality of the
company. In brief, the whole theory can be visualized as figure 2.
It is important to note that most factors discussed above which would finally influence the company, are
soft variables such as loyalty, satisfaction, value added and internal quality of company. These are not
easily captured into measurement methods used to evaluate a company. This is on of the reasons this
dynamic has been ignored for several years (Senge & Oliva).
One important characteristic of discussed theory is its being a chain of casual relationships rather than
closed loop. But an open loop can not contribute to the exponential growth of new-type companies, shown
in reference mode. This is where we would concentrate in the rest of the paper to change it into a dynamic
model.
Employee «~~. Employee Internal
Productivity + Satisfaction + Quality of
Company
+
Value
Added + Customer + Customer + _ profitablity
ZA ——— Satisfaction _ Loyalty :
Variety *
Figure 2
All theories and reasons discussed can be summarized by the following propositions:
1- The policy of putting most pressure on improving machines instead of front-line workers.
2- The cause and effect chain discussed above, which can lead to both success and failure depending on its
direction.
3- Profitability of service which lead to increasing competition, and this was a cause of better service
quality and more customer expectations.
4- Measurement methods not capable of taking soft characteristics of industry into account, resulting in
ignorance of these fundamentally important issues in service.
Focus of the research
In the rest of the paper we are going to take out cause and effect loops explaining reference mode behavior
regarding theories mentioned above. This way we would be able to make a quantitative connection
between theories and historical reference mode through building a model. A fter that we will use the model
to better understand and evaluate theories and get a better insight into the structures causing this behavior.
Structure of model
In the problem we are discussing, we can identify four important sections: total market, customers, quality
of service and company. Interaction of these parts would shape the behavior of model (Figure 3).
In the reference mode of old companies, we can see an
7 overshoot-decline behavior. The first positive loop that can
Customers lg Total Marke p, responsible for initial growth of these companies is just a
{ } common one observed in most cases of growth in economic
units. In this loop the increasing profit of company leads to
expansion and more advertisement so that they can attract
Quality of more customers. The more the customer, the bigger the
revenue and profit would be and this way the virtues circle
is completed (Figure 4, right). It is important to note that
the increasing need for service in the community (during
initial decades) and the gap between the need and presence
of service companies leads to profitability of this industry,
Company and therefore this loop is not bounded for some time. In the
. model we have shown this assumption putting a large
Figure 3 number of total customers which wouldn’t stop growth
soon. The other positive loop comes from the old paradigm
in service: the companies use and improve technology more and more, therefore they come to decrease
their variable costs and also decrease the salary they pay, leading to more profitability and growth (Figure
4, left). In reality, level of technology is bounded, so when reaching this boundary the exponential growth
driven from this loop would stop and companies insisting on it, would find it no more effective in
accelerating the growth. Any way, having reached this point, it is hard to draw back, because the company
has set equilibrium between its cost and revenue, based on low salary and low variable cost. Changing
this company wide structure into another paradigm is practically hard.
Technolo
erttiay* Level in “a
Company oN +
~ - a * 41) Revenue
i+) Percent of (+L) )
~~ Variable ~ Rookies Service A dvertisement +
2 Cost 4) -f Capacity pit
mA eae NS sin of unten
Total Cos Salary Pany
Figure 4
Now the question is: what did stop the growth of flourishing companies?
As a first idea, there is always a market boundary which can account for a companies growth being
blocked, but is it the case in our dynamic theory? For sure not all companies using old approach, had
reached their market limit, to be blocked by this factor. However, fast growth of service industry has
decreased the gap between demand and supply for service, and therefore the limited market has triggered
serious competition. As a result, while not forgetting to put this aspect in the model, we should look for
some other explanation. Historical data shows parallel growth of new companies while old ones were
declining. This indicated that old companies lost their advantage in the competition and therefore gave up
their market share to newcomers. As discussed in historical development of the theories, we can look for
the reason of this defeat in ignorance to service quality and variety, perceived by customer. Companies
paying too much attention to technology, forgot to improve their front-line workers and therefor did not
improve their quality. As far as it was the dominant approach in the industry, people wasn’t expecting
anything better. But gradually things changed and profitability of service, attracted many new competitors
with new ideas and approaches. And they put more emphasis on quality to gain better market share. This
increased the quality needed to satisfy customers leading to more pressure on old companies who were yet
trying to grow using technology. In fact ignoring employee capabilities and variety of service (which was
probably a result of too much relying on technology (Schlesinger & Heskett 1991)) kept the quality of
these companies low and stopped their growth when facing new competitors. We can change this dynamic
story into cause and effect loops in Figure 5.
an + Profitablity
a Ss
Customers 4—) +
Normal ~~ Technology
+ Variaty in 74
Market Vari i:
Customer Satisfaction arietyin +N
~s * customer , Service Rookies
perception ercen
‘ of variaty ra .
Customer
Perception i—) Service Capacity
: of Quality 4 ~~ +
Normal Quality of Service
Quality in
Market
Figure 5
In figure above, as an initial definition, we have taken the quality of service to be the ratio of service
capacity to the number of customers. Having a constant number of customers, any increase in service
capacity would improve quality. We should note that having this definition, we have to take all factors
changing quality, into service capacity. (factors like experience and productivity of workers, their liberty
to serve customer with more quality and etc.).
Drawing the structure serving for stagnation of growth in old companies, we now take our attention to fast
growth of new companies.
New successful companies tend to employ more experienced people with good communication skills.
Their employees are more satisfied than before because they are more free to serve customers as fast as
they can. On the other hand job satisfaction arisen from good quality of service would motivate these
employees more and results in their productivity. Satisfied workers, like satisfied customers, would stay
with the company much more and this means less turnover cost and more profitability. With this
approach, different parts of system would reinforce each other to make a well quality organization which
would have satisfied, loyal customers and therefore is profitable and growing (Heskett, Jones, Loveman,
Sasser & Schlesinger 1994). To capture these features in the model we design following positive loops (
Figure 6).
a’ alés Profitability Sq,
C 0 —~ Sr Personnel
ustomers +) Perentage
d ‘ustomer
Perception
of Quality
6 see
ervice Capaci
we Quality =< a
Service Quality Profitabilit
io q y\ Cc re T Cost
compan = os
Job Satisfacion (+4 Service Capacity os +)
u a emt foyes
Personnel Productivity A over Loyalty
Personnel Freedom —"™ Satisfacion—”
Figure 6
In above figure, the term “ Company Quality” refers to what we may measure by the feelings of
employees towards their jobs, colleagues, and companies. This could be improved through job training,
feedback, good colleagues and friendly atmosphere. As far as new companies invest more on job training,
hiring good personnel and satisfying their employees, their profitability would improve “company
Quality”.
Cause and effect loops discussed, are linking initial theories to a dynamic one which can be now captured
in a model. The initial model is going to show the behavior of a company run with old method. To build
the model, I took important level variables of the system and expanded the flow diagram around each,
having the structure of main loops as a guideline, then linked different parts into a complete model. This
model can be found in appendixes 1 and 2.
Here is the list of important level variables:
(Number of) Customers, (number of) Beginner and Sr. Personnel, (level of) Customer and Personnel
satisfaction, (level of) Technology and Customer and Personnel perception of quality.
Before discussing model behavior, it is desirable to clarify concept of some key variables as formulated in
the model:
Quality- In the model quality is ratio of service capacity of organization to the number of customers.
Service capacity itself is number of personnel multiplied by their productivity. And productivity is a
function of personnel satisfaction.
Satisfaction- This variable is used for both customers and personnel. For customers it is taken as a
function of quality, variety of service, and how much the customer knows the personnel of organization.
For employees this variable comes from how they find their job to he effective in satisfying customers and
how much management of company invests on improving the internal quality of organization (company
quality in figure above).
Normal quality in community- In one sector model, I had only one company. To capture the pressure this
company perceives from the side of competitors, I introduced Normal quality in community as an
exogenous time series which increases gradually as new companies enter the industry and improve the
normal quality acceptable in community. This way the important effect of competition in our dynamic
theory (which is increasing quality acceptable by customer while choosing the serving company depends
on his perception of quality, is captured. The same story is true for variable: normal variety in community.
Model behavior
The one-kind-of-company model developed, has got the following assumptions:
1- There is only one company using “old method” for decision making.
2- Price of service is set constant. ( because it played no important role in our dynamic theory.)
3- Normal quality and variety in market is set as exogenous variables which would increase gradually.
4- There is no market limit for the company.
5- Sales budget is a constant ratio of revenue.
6- Decision making rules does not change during simulation.
After running the model with these assumptions, we came up with the following behavior. The variable
“customer “ represents the number of customers who would buy service from the company and is an
indication of size of company. (Figure 7)
200,000
150,000
100,000
50,000
0
Customers - SERVIL
figure 7
Graph for Customers
1950 -:1962~=—=«974—S—«1988
‘Time (YEAR)
7998-2010
cus
This growth and decline behavior is
rather similar to our initial reference
mode. It is important to note that there
is no market limit for the company and
all stopping the growth comes from
intemal structure of the company. Of
course the exogenous variables normal
quality and variety in community
would be a key element in this
behavior. The main difference is in the
much faster decline of old companies
in this run. The reason would return to
the 6th assumption above. In this
model managers of “old companies” are deciding with the same pattern, in the other words they are not
learning from what is happening while managers in reality would learn from situation and change their
decision making rules. To explore this behavior we can look at following variables in Figure 8:
Graph for Sr Life Graph for customer satisfacion
6 2
5 1.65
4 4 13
3 a 95 me
2 6 Hh
1950 1982-1972 _——«1085-~—«*T09B «DONO 1950-1962 —~—~«TA—=—=«T8RG~C«CGGSS~C«OLO
Time (YEAR) Time (YEAR)
SrLife- SERVI1 YEAR customer satisfacion - SERVI1_ sat
Graph for Profit Percent Graph for Nor Qua Comm
2 2
2: ia = 17
0 aan 14 -
1
u
2 8
1950 1962 ‘1978-1986 «+1998 2010 = =
1950 Taso Tose ToT 1585 1965 08
Time (YEAR) Time (YEAR)
ProfitPerent- SERVIL. dal
Figure 8
Nor Qua Comm - SERVIL
do
The first graph is showing average experienced personnel loyalty to company (years staying with
company). The declining pattern indicates rising turnover costs and decreasing productivity of personnel.
The second graph shows customer satisfaction, decreasing customer satisfaction means losing customers
faster because of having less loyal customers. As a result we would need more resource allocated to sales
without having desired results and finally we are facing less profitability (third graph). It is also important
to note the effect of technology on system. As long as company is profitable, it can maintain its
technological advantage but after coming short in profit, it can no longer invest much on technology to
keep the positive loop alive. In fact investing more on technology in this situation would result in less
variety and quality level and more loss as a result, therefor there is no outcome expected, even if they
could invest on the service technology and automation: The company is trapped in its own policies and
can not improve unless it changes some of the main decision rules toward regarding customer and
personnel satisfaction more and investing on these elements rather than technological aspects. you can see
the graph for technology in figure 9. Note that 1 is taken as normal technology in industry at any time.
Graph for Technology With the insight gained from the
model, we can rephrase the
dynamic theory of our problem.
The initial approach in service
companies was to emphasize on
|_| technology. This way they could
decrease cost of service and use
less experienced employees (and
pay them less), on the other hand
people expectations in market was
0 not so high (because service
1950 1962 Loe a ie 1998 2010 industry was rather new and its
quality was of second
importance.) all these together
caused a very fast initial growth
and high profitability for these companies. This successfulness lasted for some decades, but gradually as
new comers in the industry served a higher level of quality to gain more customers. The result was
gradual increase in normal quality and variety of service acceptable in community. This meant that
market was less and less satisfied with the level of quality served by old companies. Dissatisfied
customers, not only left the companies but also made it harder to attract new ones. On the other hand,
personnel of these companies, had less and less job satisfaction when getting no positive feedback from
customers, so they became less loyal to their companies and this meant more turnover cost and less
productivity of personnel. These negative loops not only stopped the growth but also caused further
decline. Of course in reality the decline was rarely this fast and serious because managers of old
companies could perceive this trend and react on it. Changing their policies, they were able to stay in the
competition (but no more as powerful as they were before).
In the initial model, the variables normal quality and variety in community were taken as exogenous
variables increasing by time. The behavior of model was sensitive toward change of these variables. In
fact by slowing down the increasing trend of these variables seven times, we would come to a change of
behavior mode from overshoot and decline to exponential growth. This analysis reveals the need for
taking these two variables into the models dynamic instead of leaving them as exogenous variables.
This can be done by expanding the model to capture new companies and their effect on old ones. In this
case we would no more need to keep normal quality and variety in market as exogenous variables but they
can be determined from performance and competition of old and new companies. Other important
assumptions in expanding the model are stated bellow:
1- There are two companies in the model, one using old method and one using new approach.
2- Total potential customers in market are limited (9 million). This means competition to attract more
customers is present in the model.
3- The price is yet set constant.
4- The new company does not exist until 1965, when the profitability of old one would make it very
desirable to enter the industry and therefore they start working with new strategy, with rather a small size.
5- The old company reacts to the level of customer satisfaction (with some perception and decision delay)
by changing its variety and quality goal, combination of beginner to experienced personnel and improving
their companies internal quality (which accounts for personnel satisfaction).
In this model, new companies would start working with their new approach, and therefore attract
customers in the market very fast. In this way old companies lose their market share while they are not
aware of what is happening and become to an even worse position when they can not change the situation
easily even if they change their policy. figure 10 shows the behavior of model regarding number of
customers for old and new companies. (old companies customers: customers, new companies customers: _
customers2).
Technology - SERVI1_ tech
Figure 9
In this graph, the main barrier which has
_— stopped growth of new companies is market
f me web elte| limit. In other words taking out the market
\ ZL limit out of the model, new companies would
grow exponentially for ever. This difference in
behavior, comparing with first model, is due
AN to slower growth of normal quality/variety in
| /\ hk community, when we generate it internally
| A WL according to competition. Of course in reality
4 1 there would be some other barriers to growth
which might have acted before market limit in
many cases.
Another important point in this behavior, is
that old companies can not gain their initial
advantage, although they have reacted to the
situation, changing their policies (they
wouldn’t have any customers left without changing their decision variables). In the model, new
companies gain some advantage by increasing their quality and variety of service. As a result the normal
quality and variety in community increases (figure 11), this means that people are expecting more than
past and don’t get satisfied with the quality which was satisfactory 10 years ego. Therefore you should
offer much more quality service to customers for gaining their acceptance, but the companies can’t afford
much more quality because in that case the whole job is no more profitable. Totally, old companies can
not find any opportunity for improving their quality as much as needed to improve their market share ( In
the model, they could gain this change solely through improving their quality goal along with some other
parameters, but this wasn’t financially affordable.) The best choice for these companies, comes from
improving their variety of service and increasing internal quality of organization ( policy effect on
personnel life in the model). In reality this means having more new ideas, giving personnel liberty to
satisfy customers in any way they find to be appropriate and making a more friendly and satisfactory
environment in the organization.
Graph for Nor Qua Comm
Figure 10
Profit Percent of New Companies
4
t L
0 ° Se
us) aC Tao CSCO 1968 tad 1981902 ~——00T——0L
Time (YEAR) Time (YEAR)
NorQuaComm-SERM24 © dal ProftPercen2- SERM24. © dal
Figure 11
Another important policy issue from the model, arises from the profit percent for the new companies. As
you can see in figure 12, there is a declining pattern in this variable. How is it explained?
In the model we have defined quality the as ratio of quality capacity to customers and quality capacity is
product of personnel to their productivity. So, for increasing quality, we need to increase personnel or
their productivity while the later is limited. Therefore when we gradually reach the boundary of personnel
productivity through different policies, we would need more personnel to improve the quality and pay
more wage for serving the same number of customers. As a result, increasing quality has got also a
negative effect on profitability. It is important to note that in this case positive loops of relating quality to
profitability would become ineffective after some time because normal quality in community would
increase and therefor greater absolute quality (useful time of personnel spent on one customer) is regarded
as normal for customers ( having no more positive effect on their loyalty and satisfaction). In reality this
story might differ in some ways: quality perceived by customers in not only a function of time spent on
them, but also many other factors are important. We have tried to capture these features in productivity of
personnel and variety of service which are both limited in the model while innovative new ideas evolving
in the service industry are different, they can improve quality significantly while you don’t need to spend
so much on them and their variety is not bounded. The whole discussion here has got an important policy
output: all companies should be aware that no matter how good is their quality and customer satisfaction
today, it would become ordinary in a few years, so they should be always looking for new, innovative ideas
to improve their quality. This is a never ending process and profitability of company in long term is
bounded to it.
Strategy for improving the model
The first purpose of building this model has been deepening understanding of paradigm shift in service
industry. While the model is rather capable of reproducing historical mode of behavior, testing a model
validity is rather a continuos process of doing different tests on it and checking diverse policies( Forrester
& Senge, 1980), so to improve the model there are always new steps to be taken. These imperial tests
would help us understand the theory and its usefulness better and know its limitations. On the case of this
model of service industry I would suggest following points as some important ones to work on:
1- This model is not tuned with any real service industry so parameters are mostly based on intuition.
Sensitivity analysis on some important parameters shows no change in the mode of behavior but
deviations in values were significant so it is very desirable to build the model on some real parameters to
see if it is capable of reproducing historical behavior in that specific situation. In this case, it might be
desirable to change some parts of the model in order to represent a real case.
2- Most important dynamic of this model arises from some soft variables such as quality, variety,
technology, satisfaction and effects of these on each other, as a result the model stands on many lookups.
Each of these lookups can be a case of more investigation in different situations, and some of them which
are representing decision making rules, will differ according to different managers, so much work arises
from investigation of lookups to make them more valid.
3- While this model is trying to capture theories mentioned in the beginning of the paper, it is ignoring
some parts which would be important in reality, to release these assumptions, I would suggest following
cases: assumption of constant price (while it plays an important role in competition, variety and use of
technology in reality), no service backlog or rework ( in this model I have taken no backlog for service
and shortage of service capacity would result in less quality, also there is no rework), changing some
lookups into structures.
4- I have taken a simple definition for quality ( as defined in definition part). Using this definition, some
creative new ideas are missing in the model. To expand the model, putting some structure to capture these
features would be very helpful.
5- Having an improved model, it would be desirable to use it for building an interactive learning
environment which can rise many new questions while giving more insight about the dynamic of the
system to managers in the service industry.
References
Forrester, J. W and Senge P.M. 1980. Tests for Building Confidence in System Dynamics Models.
TIMS Studies in the Management Sciences, Vol. 14 (1980) : 209-228.
Heskett J.L. , Jones T.O., Loveman G.W. , Sasser W.E. , Jr. , and Schlesinger L.A. 1994, Putting the
Service-Profit Chain to Work. HARVARD BUSINESS REVIEW, Vol (March-April 1994):
164-175
Lumsdane E. and Lumsdane M. 1995. Creative problem solving, thinking skills for a changing world.
McGrawHill press.
Schlesinger L.A. and Heskett J.L. 1991, The Service-Driven Service Company. HARVARD BUSINESS
REVIEW , Vol (September: October 1991): 71-81
Schlesinger L.A. and Heskett J.L. 1991, Breaking the Cycle of Failure in Services. Sloan Management
Review, Vol. (Spring 1991): 17-28.
Senge P.M. and Oliva R. Developing a Theory of Service Quality/Service Capacity Interaction.
10
Appendix 1: Flow Diagram of the One-Sector Model:
Personnel View
sinner Personel> <r Personel> «Ratio Tech>
\ “a policy com
<Des Com> y
am Com Des Com
<Desired Personel> HEmBe NoMT EiTeG
* TiEmSr
T1Eff Tech Com
|B Emplo: ired Personel>
mt ‘ Nomnal SrLife
Mat Time .
Srlife policy effect y
Ratio Sr Life
one Sr Personel
Be Leave Maturing Sr Leave
Be Life“ BL Policy Eff
+ TIEff Sat BL
Sos BegLife Ff sat BL“ <Ratio Per Sat>-
“ ~ ~————™ Eff Sat SL
Nor Beg Life _F
Tl Eff Sat SL
Satisfaction
Aa <Ratio Var>
Normal Sat Cus7 Des Sat Cus Eff Variaty Sat Cus ;
w_T Eff Variaty Sat Cus
Ratio Cus Sat GasoRee EAT Qua cs <Ratio Cus Perc Qua>
satisfacion
Change Sat TI Eff Qua Sat Cus
Time Ch: Sat C <Ratio Per Perc Qua>
amy Change) Seb Cus Tl Eff Qua Sat Per
Normal Sat Per Des Sat pe Eff Qua Sat Peta ——
Ratio Per Sat Personnel
satisfacion
Cha Sat Per
Time change Sat Per
11
Variaty, Technology, Quality standard & Desired personnel
< ners> <Ratio Tech>—
<Quality Standard> Customers ° EffTechVar — Variaty Policy
Sag Tl sisia™
ra <Nommal Prod $r> Variaty
es C <Profit Gap> Norvacconi
<Des Com> /\. Prod Beg> gy Ratio Var
<Ave Cus Sales> <tl power>
EffPreatTech TIE Profit'Tech = 7”
Tech Life
Nor Tech Inc <Time>
Technology .
Tech Aging Tech Increase Quel Goal
a — Real Quality>
Eff Gap On Tech Inc Des Qua Stan —e—_
Ratio Tech <Nor Qua Comm>
A Tl Eff Gap Tech Inc oa
NorTech Change Standard
Time Change Standard we
Financial
<Sr Personel> Normal Profit Ind
Sr Per Sal: ~m Profit Gap
Beg Per Sal ___‘g= Salary ;
<Beginner Personel> mi ge
7 a
<Tumover Cost> a. YT
udget> ———t™ Other Exp Total Cost
Revenue
Pro Var Cost
TI Eff Tech Exp an Exp Unit
. Sales ope
Eff Tech Exp Unit <Customers> Unit Price
<Ratio Tech> Nor Exp Unit ‘Ave Cus Sales
<B Employ>
3e Leave> —————
Be Leaver Tumover Cost
<Sr Employ — 7A x.
Employing Cost
Leaving Cost
<Sr Leave>
12
Quality
<Com>
TI Eff Set Table ee p <Ratio Beg Life>
<Ratio Per Sat> _Eff Sat Prod _—~>_ ; Eff Knowing
es Quality Capacity ke | NA atatio SrLife>
Luts oN <Ave Cus Sales>
, . Real Quality Time Cus Perc
<Beginner Personel> 7
<Sr Personel> Time Fer Pere \ WA
SY \ ) Cus Perc Qua
Nor Quality Capacity Ratio Per Perc Qua
Normal Prod Sr
Nor Prod Beg Ratio Cus Perc Qua
8 power Nor Qua Comm sTime>
Tl Nor Qua
<Revenue> Customers
Sales Budge
sate Catone SEC TI Eff Sat Cus Lif
at Cus Life
Cus Inc Cus Leave
PCusExp Eff Sat Cus Life
0
NoExp Nw__Eff Sat Exp gw
ag Cus Life
<Ratio Cus Sat> <Ratio Cus Sat>
CommPercTime
Tl Eff Sat Exp
Nor Cus Life
13
Appendix 2 - Equations of One-Sector model
Ave Cus Sales = 50
~ sale/Y EAR/cus
~ average sales of a customer in one year
Eff Sat Cus Life = Tl Eff Sat Cus Life(Ratio Cus Sat)
~dmnl
~ Effect Satisfaction On Customer Life
Ratio Var = Variety/(1+(Nor Var Comm(Time)-1)*tl power)
~dmnl
~ratio of variaty
Nor Qua Comm =1+tl power*(T1 Nor Qua(Time)-1)
~dmnl
~ normal quality in community
Nor Quality Capacity = Beginner Personel*Nor Prod Beg+Sr Personel*Normal Prod Sr
~sale/YEAR
~ Normal Quality Capacity of company denying satisfaction
Ratio Per Perc Qua = DELAY 1(Ratio Cus Perc Qua,Time Per Perc)
~dmnl
~The quality as perceived by personnel
Des Sat Per = Eff Qua Sat Per*Normal Sat Per
~ sat
~ Desired Satisfaction of Personel
Mat Time = NoMT
~YEAR
~ average time it takes a rooky to become experienced
tl power =1
B Employ = (Desired Personel* (Des C om/(Des Com+1))-Beginner Personel)/TiEmBe+Be Leave
~ person/Y EAR
~ Beginers employment rate
Be Leave = Beginner Personel/Be Life
~ person/Y EAR
~ rate of leaving of beginners
Be Life = BL Policy Eff*Eff Sat BL*Nor Beg Life
~YEAR
~ actual beginers life
Beg Per Sal = 12000
~$/YEAR
~ beginner personnel salary
Beginner Personel = INTEG (B Employ-Be Leave-Maturing, 15)
~ person.
BL Policy Eff = GAME(1)
~dmnl
~ policy of the company to keep its beginer personel
Cha Sat Per = (Des Sat Per-Personnel satisfacion)/Time change Sat Per
~sat/YEAR
~rate of changing personnel satisfaction
Change Sat = (Des Sat Cus-customer satisfacion)/Time Change Sat Cus
~sat/YEAR
~ rate of changing satisfaction of customer
Change Standard = (Des Qua Stan-Quality Standard)/Time Change Standard
~1/YEAR
~rate of changing quality standard
Com = Beginner Personel/Sr Personel
~dmnl
14
~ the combination of personel in the system
CommPercT ime = 1.5
~YEAR
~ Community Perception of Satisfacion Time
Cus Inc = Sales Budget/PC usExp
~cus/Y EAR
Cus Leave = Customers/Cus Life
~cus/Y EAR
~ rate of customers leaving the company
Cus Life = Eff Sat Cus Life*Nor Cus Life
~YEAR
Cus Perc Qua = SMOOTH (Real Quality, Time Cus Perc)
~NOR PER/cus
~ Customer Perception of Quality
customer satisfacion = INTEG (Change Sat, 1)
~sat
Customers = INTEG (Cus Inc-C us Leave, 1000)
~cus
~ number of customers
Des Com = Eff Tech Com* policy com
~dmnl
~ Desired combination of Beginer customer to total customer
Des Qua Stan = IF THEN ELSE (Quality G oal>=Real Quality,(Nor Qua Comm+4Q uality G oal+Real
Quality)/3, (Quality Goal
+Nor Qua Comm)/2)
~dmnl
~ desired quality standard in company
Des Sat Cus = Eff Qua Sat Cus*Eff Variaty Sat Cus*Normal Sat Cus
~ sat
~ Desired Satisfaction of customer
Desired Personel = Quality Standard*C ustomers* Ave Cus Sales/(Normal Prod Sr/(1+Des Com)+Nor Prod Beg
*Des Com/(1+Des Com))
~ person.
~ desired number of personnel
Eff Gap On Tech Inc = T] Eff Gap Tech Inc(Ratio Tech)
~dmnl
~ effect of gap between company technology and community technology on increase of technology
Eff Knowing =Table Eff Know (Ratio Beg Life*C om/(1+C om)+Ratio Sr Life/(1+Com))
~dmnl
~ Effect of knowing personel on quality perceived by customer
Eff Profit Tech = T] Eff Profit Tech(Profit Gap)
~dmnl
~ effect of profit on technology table
Eff Qua Sat Cus = Tl Eff Qua Sat Cus(Ratio Cus Perc Qua)
~dmnl
~ effect of quality on satisfaction of customer
Eff Qua Sat Per = Tl Eff Qua Sat Per(Ratio Per Perc Qua)
~dmnl
~ Effect of quality on satisfaction of personnel
Eff Sat BL = Tl Eff Sat BL (Ratio Per Sat)
~dmnl
~ Effect of Satisfaction On Beginers Life
Eff Sat Exp = T] Eff Sat Exp(SMOOTH Ratio Cus Sat,CommPercTime))
~dmnl
~ Effect of Satisfacion on Expenses for Advertisment
Eff Sat Prod = T1 Eff Sat Prod(Ratio Per Sat)
~dmnl
~ Effect of satisfation of Productivity of perssonel
15
Eff Sat SL = T] Eff Sat SL (Ratio Per Sat)
~dmnl
~ Effect of Satisfacion on Sr Life
Eff Tech Com =T] Eff Tech Com(Ratio Tech)
~dmnl
~ Effect of Technology on Combination
Eff Tech Exp Unit = T] Eff Tech Exp Unit(Ratio Tech)
~dmnl
Eff Tech Var = T] Eff Tech Var(Ratio Tech)
~dmnl
~ effect of technology on variaty of company products
Eff Variaty Sat Cus = Tl Eff Variaty Sat Cus(Ratio Var)
~dmnl
~ Effect of variaty on satisfaction of customer
Employing Cost = 1000
~ $/person
~ cost of employing one new personnel
Exp Unit = Eff Tech Exp Unit*Nor Exp Unit
~dmnl
~ expenses on on unit
Fixed Cost = Other Exp+Salary
~$/YEAR
Leaving Cost = 500
~ $/person
~ cost of personnel leaving the company
Maturing = Beginner Personel/Mat Time
~ person/Y EAR
~ the rate of maturing of rookies
NoExp = 150
~ $/cus
~ normal budget used to attract a new customer
NoMT =3
~YEAR
~ normal maturing time
Nor Beg Life = 2
~YEAR
Nor Cus Life = 3
~YEAR
Nor Exp Unit =8
~ $/sale
~ normal expenses for one unit of product
Nor Prod Beg = 1600
~ sale/person/Y EAR
~ normal number of sales a beginer can handle in one year
Nor Tech = 1
~tech
Nor Tech Inc = 0.07
~1/YEAR
~ normal rate of increasing technology
Nor Var Comm _ ([(1950,0)-(2010,2)],(1950,1),(1971.65,1.01408)
,(1983.09, 1.07746), (1995.15, 1.19718), (2004.12, 1.3169), (2010,1.4507)
)
~var
Normal Prod Sr = 2700
~ sale/(Y EAR*person)
~normal number of sales an Sr. can handle in one year
Normal Profit Ind = 0.08
~dmnl
16
~ normal profit percent in industry
Normal Sat Cus = 1
~ sat
~ Normal Satisfaction of Customer
Normal Sat Per = 1
~ sat
~ Normal Satisfaction of Personel
Normal Sr Life = 4
~YEAR
Other Exp = Sales Budget+Turnover Cost
~$/YEAR
PCusExp = Eff Sat Exp*NoExp
~ $/cus
~ Per customer Expenses for Advertisment
Per Perc Qua = SMOOTH (Real Quality, Time Per Perc)
~NOR PER/cus
~ Personel perception of quality
Personnel satisfacion = INTEG (Cha Sat Per,1)
~ sat
policy com = GAME(2)
~dmnl
~ Policy for the ratio of Beginers to Sr.s
Pro Var Cost = Exp Unit*Sales
~$/YEAR
~ the variable cost related to product
Profit = Revenue-T otal Cost
~$/YEAR
Profit Gap = Profit Percent-Normal Profit Ind
~dmnl
Profit Percent = Profit/Revenue
~dmnl
Quality Capacity = Eff Sat Prod* Nor Quality Capacity
~sale/Y EAR
~ the capacity of the company for offering service ( in number of sales capable of handling)
Quality Goal = GAME(1)
~dmnl
~ the goal of company for ratio of quality capacity to customers need
Quality Standard = INTEG (Change Standard,1)
~dmnl
Ratio Beg Life = Be Life/Nor Beg Life
~dmnl
Ratio Cus Perc Qua = Cus Perc Qua/(Nor Qua Comm)
~dmnl
~ ration of Customer perception of quality to normal quality in community
Ratio Cus Sat = customer satisfacion/Normal Sat Cus
~dmnl
~ratio of customer satisfaction to normal
Ratio Per Sat = Personnel satisfacion/Normal Sat Per
~dmnl
Ratio Sr Life = Sr Life/Normal Sr Life
~dmnl
Ratio Tech = Technology/Nor Tech
~dmnl
Real Quality = Quality Capacity/(Ave Cus Sales*C ustomers)* Eff Knowing
~dmnl
~ Real Quality of company for customers
Revenue = Sales* Unit Price
~$/YEAR
17
Sal Per = 0.1
~dmnl
~ration of total sales used for adv.
Salary = Beginner Personel*Beg Per Sal+Sr Personel*Sr Per Sal
~$/YEAR
~ expenses for salary
Sales = Customers*A ve Cus Sales
~sale/YEAR
Sales Budget = Sal Per*Revenue
~$/YEAR
Sr Employ = (Desired Personel/(1+Des Com)-Sr Personel)/TiEmSr+Sr Leave
~ person/Y EAR
~ Rate of Sr. Employment
Sr Leave = Sr Personel/Sr Life
~ person/Y EAR
~rate of Sr. personel leaving the system
Sr Life = Eff Sat SL*Normal Sr Life*Sr life policy effect
~YEAR
~actual Sr. Life
Sr life policy effect = GAME (1)
~dmnl
~ policy of company to keep its Sr. personel
Sr Per Sal = 20000
~$/YEAR
~ Sr. Personnel salary
Sr Personel = INTEG (Maturing-Sr Leavet+Sr Employ,10)
~ person.
~ number of experienced personel
Table Eff Know _([(0,0)-(2,2)],(0,0.838028), (0.360825,0.887324)
,(0.721649,0.922535), (1,1), (1.29897, 1.07746), (1.62371,1.12676)
,(1.99485,1.14085) )
~dmnl
~ Effect of knowing personel on quality table
Tech Aging = Technology/T ech Life
~tech/Y EAR
~rate of aging of technology
Tech Increase = Eff Gap On Tech Inc*Nor Tech Inc*T echnology* Eff Profit Tech
~tech/Y EAR
Tech Life = 15
~YEAR
~ life of technology
Technology = INTEG (+Tech Increase-Tech A ging,1)
~tech
TiEmBe = 0.1
~YEAR
~ time to employ a beginer
TiEmSr = 0.2
~YEAR
~ time to emply a sr
Time Change Sat Cus = 0.5
~YEAR
~ time it takes to change satisfation of customers
Time change Sat Per = 0.3
~YEAR
~ time it takes to change the satisfaction of personnel
Time Change Standard = 0.5
~YEAR
Time Cus Perc = 0.8
18
~YEAR
~ time for customer to perceive real quality
Time Per Perc = 0.5
~YEAR
~ time for personel to perceive real quality
TIEff Gap Tech Inc ({(0,0)-(2,2)],(0,1.2),(0.262887,1.17606)
(0.706186, 1.11972), (0.891753, 1.06338), (1,1),(1.1701,0.753521)
,(1.37113,0.443662), (1.55155,0.253521), (1.74227,0.0915493), (1.88144, 0.0422535)
(2,0) )
~dmnl
~ effect of technology gap on technology increase
TIEff Profit Tech ({(-0.1,0)-(0.1,2)], (-0.1,0.1),(-0.0881443,0.415493)
,(-0.0737113,0.647887), (-0.0510309, 0.788732), (-0.0329897,0.84507)
,(-0.0175258, 0.894366), (0,1),(0.0190722, 1.14085), (0.0458763, 1.32394)
,(0.0680412, 1.40845), (0.1,1.5) )
~dmnl
~ effect of profit on technology increase table
TIEff Qua Sat Cus ({(0,0)-(2,2)],(0,0.2),(0.278351,0.316901)
,(0.489691,0.492958), (0.695876,0.71831),(1,1),(1.3299, 1.24648)
)(1.57732, 1.41549), (1.76804, 1.52113), (2,1.6) )
~dmnl
~ Effect quality on satisfaction of customer table
TIEff Qua Sat Per ([(0,0)-(2,2)],(0,0.7),(0.28866,0.71831)
,(0.494845,0.767606), (0.675258,0.838028), (1,1), (1.30412,1.14789)
(1.55155, 1.22535), (1.78866, 1.28169), (2,1.3) )
~dmnl
~ Effect quality on satisfaction of personnel table
TIEff Sat BL _ ({(0,0)-(2,6)], (0,0.5), (0.479381,0.591549),
(0.778351,0.78169), (1,1), (1.2732, 1.47887), (1.53608,2.07042),
(1.70103,2.38732), (1.85052,2.61972),(1.92784,2.72535), (1.98969,2.74648)
)
~dmnl
~ Effect of Satisfation of Beginers Life Table
TIEff Sat Cus Life ({(0,0)-(2,5)],(0,0.1),(0.345361,0.28169)
,(0.628866,0.528169), (0.850515,0.774648), (1,1), (1.19072, 1.25)
,(1.35052, 1.5669), (1.53093,2.28873),(1.65979,3.20423), (1.82474,3.85563)
,(1.91237,3.94366), (2,4) )
~dmnl
~ Effect Satisfation On Custumer Life Table
TIEff Sat Exp ([(0,0)-(2,10)],(0,8),(0.190722,5.66901),(0.386598,3.90845)
,(0.592784,2.28873), (0.768041, 1.47887), (0.860825, 1.19718),(1,1)
,(1.30412,0.669014), (2,0.4) )
~dmnl
~ Effect Satisfaction on Expenses Table
TIEff Sat Prod ([(0,0)-(2,2)],(0,0.3),(0.340206,0.394366),
(0.675258,0.697183), (1,1), (1.41237, 1.21831), (2,1.4) )
~dmnl
~ Effect of Satisfaction of Productivity of personel Table
TIEff Sat SL ({(0,0)-(2,6)], (0,0.2), (0.42268,0.316901),(0.675258,0.507042)
,(0.912371,0.78169), (1,1), (1.15979, 1.45775), (1.34536,2.1338)
;(1.59794,3.46479), (1.76804, 4.56338), (1.81959, 4.8169), (1.87629,4.94366)
,(1.92784,5.00704),(2,5) )
~dmnl
~ Effect of Satisfacion On Sr. Life
TIEff Tech Com ([(0,0)-(2,2)], (0,0.8), (0.304124,0.802817),
(0.551546,0.84507), (0.819588,0.922535), (1,1), (1.19072,1.1831)
(1.31959, 1.28873), (1.45876, 1.39437), (1.61856, 1.42958), (2,1.5)
)
19
~dmnl
~ Effect technology on combination Table
TIEff Tech Exp Unit ({(0,0)-(2,2)],(0,1.3), (0.247423, 1.28169)
,(0.520619, 1.22535), (0.78866, 1.1338), (1,1),(1.19072,0.894366)
,(1.5,0.788732), (1.74227,0.725352),(2,0.7) )
~dmnl
TIEff Tech Var ([(0,0)-(2,2)],(0.0103093, 1.0493), (0.293814, 1.04225)
,(0.71134,1.02817), (0.876289, 1.02113), (1,1), (1.08763,0.950704)
,(1.17526,0.84507), (1.40206,0.669014), (1.62887,0.577465), (1.78866,0.528169)
1(2,0.5) )
~dmnl
~ effect of technology on variaty table
TIEff Variaty Sat Cus ([(0,0)-(2,2)], (0,0.4),(0.247423,0.478873)
,(0.489691,0.612676), (0.690722,0.739437), (0.860825,0.887324)
(1,1), (1.15979, 1.11268), (1.37113, 1.19718), (1.60309, 1.24648)
,(1.82474,1.27465), (2,1.3) )
~dmnl
~ Effect variaty on satisfaction of customer table
TINor Qua _({(1950,0)-(2020,2)], (1950, 1), (1965.52, 1.0493)
,(1971.65, 1.08451), (1980.31, 1.1338),(1989.15, 1.21127), (2001.6,1.35915)
(2010.98, 1.48592), (2020, 1.67606) )
~NOR PER/cus
Total Cost = Fixed Cost+Pro Var Cost
~$/YEAR
Turnover Cost = Employing Cost*(B Employ+Sr Employ)+L eaving C ost* (Be Leave+Sr Leave)
~$/YEAR
~ turnover cost of company
Unit Price = 20
~$
Variaty = Variaty Policy*Eff Tech Var
~var
~variaty of products of company
Variaty Policy = GAME(1)
~var
~ policy of company for variaty of products
AEG CR AK
Control
SECA GGA KCK
Simulation Control Paramaters
FINAL TIME =2010
~YEAR
~The final time for the simulation.
INITIAL TIME = 1950
~YEAR
~ The initial time for the simulation.
SAVEPER =
TIME STEP
~YEAR
~The frequency with which output is stored.
TIME STEP =0.1
~YEAR
~The time step for the simulation.
20
CC BY-NC-SA 4.0