Construction, test, analysis, and adjustment of the Basic System Dynamics Model.29

This research mainly applied the following resources to set up the systematic dynamics model; The first one is the materials in class from Dr.Tu. “the systematic simulation and dynamic decision”. The second one is “Corporate Planning and Policy Design: A System Dynamics Approach”, Lyneis, 1980. The method uesd in this research is to meet and cooperate with the structure of the performance

measurement index in balanced scorecard, and then apply the target company into this research.

This research uses 4 system dynamics basic flow wire models as the basis of the simulation models, and then carry out the model constructing.

(1) The causality of action taken to incorporate the strategic goal and performance measurement indicators to the company policy.

(2) The reference model of carrying out the strategic goal and performance measurement indicators to the action taken for company policy.

(3) The basic flow wire models according to the four perspectives of balanced scorecard.

(4) The feedback loop reference models that include balanced scorecard and the strategic goal and the action taken for company policy.

3.3.1 Setting-up the Foundation of the Systematic Dynamics Model

Systematic dynamics is mainly built and constructed in the elements, relationship and their combination. And form the systematic dynamics model through the interactive mechanism.

Main elements of the structure of systematic dynamics are:

(1) Level : Indicate the thing or material in the real world that accumulate with elapsed time. It represents the system state at the specific time point. It takes shape and has been accumulated by inflow and outflow after a period of time.

(2) Rate : Show a certain speed of flow. Namely the flow quatity in unit of time. It directly determines the control point that measures the level state. It is the starting point of making policy and taking action too.

(3) Auxiliary variable : Represent three kinds of meanings: Process of information management, some specific environmental parameter values (constant ), systematic test functions or values. The first two can be regarded as a part of quantity of the rate, while the last one is suitable for testing all kinds of situation of the model behaviors.

(4) Flow wire: Mean the entity flow or information flow in the whole loop. It is mainly used for transmitting the information in the exchange system and makes the structure of the whole system intact to fulfill the information feedback purpose.

In order to meet the cosistency of model constructing, it is necessary to consider the unit of simulation time. it can be second, minute, day, week, month, year. It is not only an important factor for systematic dynamics simulation modeling process, but also important to the result of model analysing and testing.

3.3.2 Basic Systematic Dynamics Model for Balanced Scorecard 3.3.2.1 Fundamental modes of dynamic behavior-- Goal Seeking Mode

Change takes many forms, and the variety of dynamics around us is astounding. You might imagine that there must be a correspondingly huge variety of different feedback structures to account for such a rich array of dynamics. In fact, most dynamics are instances of a fairly small number of distinct patterns of behavior, such as exponential growth or oscillation.

The most fundamental modes of behavior are exponential growth, goal seeking, and oscillation. Each of these is generated by a simple feedback structure: growth arises from positive feedback, goal seeking arises from negative feedback, and oscillation arises from negative feedback with time delays in the loop. Other common modes of behavior, including S-shaped growth, S-shaped growth with overshoot and oscillation, and overshoot and collapse, arise from nonlinear interactions of the fundamental feedback structures.

Positive feedback loops generate growth, amplify deviations, and reinforce change.

Negative loops seek balance, equilibrium, and stasis. Negative feedback loops act to bring the state of the system in line with a goal or desired state. They counteract any disturbances that move the state of the system away from the goal. All negative feedback loops have the structure shown in Figure 3.3. The state of the system is compared to the goal. If there is a discrepancy between the desired and actual state, corrective action is initiated to bring the state of the system back in line with the goal.

When you are hungry, you eat, satisfying your hunger; when tired, you sleep, restoring your energy and alertness. When a firm’s inventory drops below the stock required to provide good service and selection, production increases until inventory is once again sufficient.

Every negative loop includes a process to compare the desired and actual conditions and take corrective action. Sometimes the desired state of the system and corrective action are explicit and under the control of a decision maker (e.g., the desired level of

inventory). Sometimes the goal is implicit and not under conscious control, or under the control of human agency at all. The amount of sleep you need to feel well rested is a physiological factor not under your conscious control. The equilibrium surface temperature of the earth depends on the flux of solar energy and the concentration of greenhouse gases in the atmosphere, among other physical parameters. And a cup of coffee cools via negative feedback until it reaches room temperature.

In most cases, the rate at which the state of the system approaches its goal diminishes as the discrepancy falls. We do not often observe a constant rate of approach that suddenly stops just as the goal is reached. The gradual approach arises because large gaps between desired and actual states tend to generate large responses, while small gaps tend to elicit small responses. The flow of heat from your coffee cup to the air in the room is larger when the temperature gap between them is large and diminishes as the gap falls. When coffee and room temperatures are equal, there is no net heat flow between them.

When the relationship between the size of the gap and the corrective action is linear, the rate of adjustment is exactly proportional to the size of the gap and the resulting goal-seeking behavior is exponential decay. As the gap falls, so too does the

adjustment rate. And just as exponential growth is characterized by its doubling time, pure exponential decay is characterized by its half life-the time it takes for half the remaining gap to be eliminated. Figure below shows examples of goal-seeking behavior.

Figure 3.3 System dynamics behavior mode -Goal Seeking Mode Source: Sterman, Business Dynamics, 2003

State of the

3.3.2.2 The reference model of strategic goal, Performance Measurement Item and Improvement action

Performance measurement Item could be a level variable or a calculated value from an auxiliary variable. By increasing or decreasing the rate variable, adjust the volume of level. As for strategic goal, discrepancy, and policy action, it has to use auxiliary variable to calculate and set up its value. So that it will feedback to rate variable and change the level, the current performance measurement Item condition. The following figure is showing the relationship of these variables.

Performance Measurement

Item Performance

Decrease Performance

Increase

Performance Improvement Effect Performance

Indicator Goal

Discrapancy (GAP)

Performance Improvement Action

Figure 3.4 The Reference Model of Strategic Goal, Performance Measurement Item and Policy (Improvement) Action

Source: Chao-jen Huang, A study on Interaction mechanism between BSC strategy goal and Performance Index, NSYSY, 2005

3.3.2.3 The reference model for BSC with negative feedback loop

The following figure indicates the basic system dynamics model for Balanced Scorecard. It could be used to help enterprises construct the basic structure of the model for evaluating performance measurement.

Operating Income

Customer Satisfaction

Internal Performance

Employee Efficiency Customer2Income

Internal2Customer

Efficiency2Intenal

Employee Efficiency Promotive Action Internal Performance Promotive Action Customer Satisfaction

Promotive Action

Strategy Goal

KPI-1

KPI-2

KPI-3 Cost & Expense

GAP-1

GAP-2

GAP-3

GAP-4

Actions Taken

Figure 3.5 The Basic Flow Wire for the Four Perspectives of BSC

Source: Chao-jen Huang, A study on Interaction mechanism between BSC strategy goal and Performance Index, NSYSY, 2005

3.3.3 Test, Analysis, and Policy Design of the Systematic Dynamics Model Systematic dynamics modelers have developed a wide of variety of specific test to uncover flaws and improve models. Model user must critically assess the model’s boundary, time horizon, and level of aggregation light of their purpose. Therefore, some questions are raised to check the model. What is the purpose of the model?

What is the boundary of the model? What is the time horizon relevant to the problem?

Does the model include the factors that may change significantly over the time horizon as endogenous elements? Is the level of aggregation with the purpose?

Sterman (2000) indicates that the question facing the modelers is never whether a model is true but whether it is useful. The choice is never whether to use the model.

The only choice is which model to use. Selecting the appropriate model is always a value judgment to be made by reference to the purpose. Without a clear

understanding of the purpose for which the model is to be used, it is impossible to whether you should use it as a basis for action.

To avoid the problems with the model, such as violating basic physical law, narrow boundary cutting critical feedbacks, modelers kept the assumption hidden from the readers, and modelers failed to include important stakeholders in the modeling process, you must insist on the highest standard of documentation. The model must be fully replicable and available for critical review. Use the documentation to assess the adequacy of the model boundary and appropriateness of its underlying

assumptions about the physical structure of the system and the decision-making behavior of people acting within it. Consider extreme condition tests and sensitivity to alternate assumptions, including assumptions about model boundary and structure, not only sensitivity to variations in parameter values.

No one test is adequate. These tests help you understand the robustness and

limitations of your model. They involve direct inspection of equations and simulations of the whole model, the assessment of historical fit, and behavior under extreme conditions. Test the robustness of your conclusions to uncertainty in your

assumptions. While parametric sensitivity testing is important, model results are usually far more sensitive to assumptions about model boundary, level of aggregation, and representation of decision-making.

Testing has been carried on right after the systematic dynamics model constructed.

By observing the testing results, analyze and describe the system behavior and consequences. And then discuss the strategic goal design and adjust the relevant performance indicators. The systematic dynamics model must be able to inspect and correct the whole process through repeated simulation. On one hand, it helps top

management or strategic planners to deeply understand the complicated behavior of the system. On the other hand, it can analyze the procedure and verify the behavior between the results and system structure through rigorous logical reasoning, and then improve the system validity to the certain degree.

Chapter 4 Case Study – AIDC Commercialized Maintenance Program

4.1 Introduction of the Target Company