Representing Qualitative Variables and Their Interactions With Fuzzy Logic in System Dynamics Modeling

We propose a method for incorporating multiple linguistic or soft variables into a system dynamics framework. A simple example is used to illustrate the procedures necessary to define linguistic variables using triangular membership functions within the VENSIM Simulation Environment. We illustrate the operations of linguistic variables through a sales and service model where two linguistic variables, i.e. customer's satisfaction with respect to service, and lead time associated with a product, impact the conversion of potential customers into customers. After having created fuzzy triangular membership functions, we obtain the combined effect of the two linguistic variables using the max-min inference procedure. For defuzzification we use the notion of the largest of maximum to translate the fuzzy representation of the combined effect into a crisp value. Finally, we provide simulation results pertaining to the probability of generating new customers and profits by considering pessimistic, optimistic and intermediate fuzzy rules for our model. Copyright (C) 2010 John Wiley & Sons, Ltd.