| The research conducted within this dissertation focuses on the development and analysis of optimization models for the location of vehicles in emergency medical service (EMS) systems. Two separate classes of problems are addressed, termed the static deployment problem and the fluid deployment problem. Within the static problem, each ambulance is assigned a “home” base. After providing service to a call, the ambulance will always return to its home base, regardless of the status of other vehicles. In contrast, within the fluid problem, vehicles are not assigned a single home base. Rather, they will be relocated from one base to another, depending on the current status of the system, in order to maintain optimal response capability at all times. Within both of these different veins of research, the dissertation begins by examining existing models that may prove relevant. In each case, however, it is determined that the existing models, while capable of determining relatively effective location plans, tend to generate inaccurate estimations of system reliability. This represents a significant problem for practical application, for if planners are to make use of these approaches, they must be able to rely upon the accuracy of the estimations of system performance prior to implementation. Given this difficulty, new approaches for optimizing both the static location problem and the fluid location problem are introduced. These new approaches integrate a simulation routine together with an optimization search strategy. In a series of detailed tests, it is demonstrated that each of the new approaches is highly successful. Based on the output of the embedded simulation routines, the predictions of system performance prove to be highly accurate. At the same time, the quality of the solutions generated by the search strategies are shown to be as good as or better than that of the solutions determined by the prior models. |
