Metodologia para dimensionamento e análise de serviços de atendimento de emergência

Abstract

Since the 70s there are studies that shows the existence of a direct correlation between the response time (time from when the call is first dispatched until the ambulance reaches the scene) of emergency medical systems and the probability of survival for victims involved in accidents. It is clear that any tool capable of providing an operational improvement for this type of system regarding the response time is very valuable to managers and the general population. This study aims to develop a methodology based on simulation optimization with the goal of dimensioning emergency medical systems. The methodology is based on three main aspects: a module based on discrete event simulation techniques for analyzing all the configurations; new procedures based on the meta-heuristics tabu search and simulated annealing for generating different scenarios; and the use of multiple regression metamodels to act as a filter and accelerate the process of simulation optimization. The basic premise to be verified is that the simulation optimization process can be accelerated and generate better results when integrated with the use of metamodels. This procedure has never been used in the context of emergency medical systems. The proposed methodology solves the ambulance location problem and the dimensioning problem, determining the optimal amount of units in each base. Two different types of ambulances were considered: basic life support unit, to respond to less severe occurrences and advanced life support unit, to respond to more severe occurrences. The optimization module is based on minimizing costs of opening bases and acquisition of ambulances. The restrictions consider a maximum response time for the system and a limit of ambulances allocated to each base. The proposed methodology is validated using data from the Emergency Medical System of the city of Belo Horizonte (Minas Gerais) and tests were conducted to identify the quality of the optimization procedures. The proposed procedures were compared with OptQuest optimizer from OptTek Systems, integrated with the simulator Arena from Rockwell Automation Technologies. In all scenarios analyzed, the best configuration provided by the proposed methodology outperforms the best configuration provided by OptQuest in terms of cost, reaching the same level of response time with less opened bases and ambulances.