Modeling uncertainties based on data mining approach in emergency service resource allocation

Abstract

Thousands of victims and millions of people are affected by natural and non-natural incidents every year. The Resource Allocation Problem can be often considered as part of a post-incident and pre-incident measure. In this paper, a new mathematical model is presented based on data mining methods to maximize the coverage of demand points by ambulances and rescue cars under conditions of uncertainty by rescue and relief (RAR) stations. Also, to increase the accuracy and quality of the results, data mining methods were used to estimate the critical parameter which is defined as “the minimum required number of each kind of equipment to cover demand points under certain conditions” based on historical data in Iran which this method is the original contribution of this article.The objectives of this research are divided into three main sections:1.Introducing a new model based on data mining for the allocation of relief resources.2.To maximize coverage of demand points.3.Final testing of the model based on real data by improving the performance, accuracy and quality of the ultimate response of the rescue organization and providing an appropriate allocation plan.The results show that the proposed model has achieved a 5.19% improvement in optimal solution over the previous model for the same data set. Also, the number of covered demand points by the ambulances improved by 7.14% in the first and second scenarios and 17.64% for the rescue cars in the second scenario.The novelty of this study is to present a new model for the allocation of relief equipment to maximize coverage of demand points based on data mining approach under uncertainty.