Discuss different steps associated with model development. What are simulation models ?

Discuss different steps associated with model development. What are simulation models ?

MODEL BUILDING/ MODEL DEVELOPMENT
The approach sued for model building or model development for managerial decision-making will vary from one situation to another. However, we can enumerate a number of generalized steps which can be considered as being common to most modeling efforts. The steps are:
1) Identifying and formulating the decision problem.
2) Identifying the objective(s) of the decision maker(s)
3) System elements identification and block building.
4) Determining the relevance of different aspects of the system.
5) Choosing and evaluating a model form.
6) Model calibration.
7) Implementation.

The decision problem for which the researcher intends to develop a model needs to be identified and formulated properly. Precise problem formulation can lead one to the right type of solution methodology. This process can require a fair amount of effort. Improper identification of the problem can lead to solutions for problems which either do not exist or are not important enough. A classic illustration of this is the case of a manager stating that the cause of bad performance of his company was the costing system being followed. A careful analysis of the situation by a consultant indicated that the actual problem lay elsewhere, i.e., the improper product-mix being produced by the company. One can easily see here the radically different solutions/models which could emerge for the rather different identifications of the problem!
The problem identification is accompanied by understanding the decision process and the objective(s) of the decision maker (s).Very often, specially in case of complex problems, one may run into situations of multiple conflicting objectives. Determination of the dominant objective, trading-off between the objectives, and weighting the objectives could be some ways of taking care or this problem. The typical objectives which could feature in such models can be maximizing profits, minimizing costs, maximizing employment, maximizing welfare, and so on.
The next major step in model building is description of the system in terms of blocks. Each of the blocks is a part of the system which has a few input variables. The decision-making system as a whole can be described in terms of interconnections between blocks and can be represented pictorially as a simple block diagram ( please refer figure4.) However, one should continuously question the relevance of the different blocks vis-à-vis the problem definition and the objectives. Inclusion of the not so relevant segments in the model increases the model complexity and solution effort.

A number of alternative modeling forms or structures may be possible. For instance, in modeling marketing decision-making situations, one may ask questions such as whether the model justifies assumptions of linearity, non-linearity (but linearizable ) and so on. Depending upon the modeling situation one may recommend the appropriate modeling form. The model selection should be made considering its appropriateness for the situation. One could evaluate it on criteria like theoretical soundness, goodness of fit to the historical data, and possibility of producing decisions which are acceptable in the given context.

SIMULATION MODELS
Simulation models are a distinct class of quantitative models, usually computer based, which are found to be of use for complex decision problems. The term’ simulation’ is used to describe a procedure of establishing a model and deriving a solution numerically. A series of trial and error experiments are conducted on the model to predict the behavior of the system over a period of time. In this way the operation of the real system can be replicated. This is also a technique which is used for decision-making under conditions of uncertainty. Generally, simulation is used for modeling in conditions where mathematical formulation and solution of model are not feasible. This methodology has been used in numerous types of decision problems ranging from queuing and inventory management to energy policy modeling. A detailed discussion of simulation is beyond the purview of this unit. For those of you who would like to read more on this, we would recommend a comprehensive text like Gordon (1987) or Shenoy et., al. (1983).