A simulation model for evaluating a set of emergency vehicle base locations: development, validation, and usage.

This paper describes our experiences in developing a simulation model for evaluating a set of emergency response vehicle base locations. The project was undertaken jointly by the University of Arizona and the Tucson Fire Department. The issues of model development, data collection, model validation, and experimentation are discussed. The critical nature of the problem and the clients' lack of experience with mathematical models, made model validation the major step in gaining user acceptance. We show that looking solely at standard performance statistics such as the calls successfully serviced, may lead to the acceptance of an invalid model. We also show that the high level of detail used in many simulation models for evaluating base locations is unnecessary in the current case. An analysis evaluating two alternative sets of locations for the Tucson system is discussed.