A stochastic multi-objective model based on the classical optimal search model for searching for the people who are lost in response stage of earthquake

Although after an earthquake the injured person should be equipped with food, shelter and hygiene activities, before anything must be searched and rescued. But disaster management (DM) has focused heavily on emergency logistics and developing an effective strategy for search operations has been largely ignored. In this study, we suggest a stochastic multi-objective optimization model to allocate resource and time for searching the individuals who are trapped in disaster regions. Since in disaster conditions the majority of information is uncertain, our model assumes ambiguity for the locations where the missing people may exist. Fortunately, the suggested model fits nicely into the structure of the classical optimal search model. Hence, we use a stochastic dynamic programming approach to solve this problem. On the other hand, through a computational experiment, we have observed that this model needs heavy computation. Therefore, we reformulate the suggested search model as a multi-criteria decision making (MCDM) problem and employ two efficient MCDM techniques, i.e. TOPSIS and COPRAS to tackle this ranking problem. Consequently, the computational effort is decreased significantly and a promising solution is produced.