Location-Routing Problem In Optimization Of Emergency Logistics System For Public Incidents

With the rapid development of socio-economic, the problems of population, resource, environment, and public sanitation are becoming worsening. When public incident happen, if we did not take effective measures to deal with it in time, the consequence could be disastrous. Dealing with unexpected public incident is a complicated systematic project, in which emergency material supplies and people's life security are the major issues that can maintain social stability. Therefore, how to optimize emergency material inventory and emergency distribution system scientifically and how to strengthen the emergency response capabilities of the whole logistics of system, are the key issues to enhance stability and reliability timelines of a public emergency early-warning defense system.Public emergency logistics system must meet the logistics needs under constraints of time, space and resources in order to achieve maximize the time effectiveness and minimize losses. Emergency logistics optimization systems can effectively reduce logistics costs and improve the efficiency of logistics services. Location-Allocation Problem (LAP) and Vehicle-Routing Problem (VRP) are two key problems in public emergency logistics system optimization. As the two is so closely linked that there is a need for further analysis of their relations, which comes to be Location-Routing Problem (LRP). At present, there are many literatures in the LRP research, but they are only for the general logistics, we have not yet found any literature of LRP in emergency logistics.In this study, we firstly reviewed and summarized the research status in emergency management, emergency logistics system and LRP. On the basis of current studies, a single-objective mathematical model of LRP was developed to maximize the time satisfaction of emergency relief in a situation of network congestion. Considering the fuzzy demand, we also developed a multi-objective mathematical model of LRP to maximize the time satisfaction and to minimize the system costs. As the LRP is a nondeterministic polynomial-hard (NP-Hard) problem, it can't be solved by using accurate algorithm. In this paper, a two-phrase heuristic algorithm and a mixed genetic algorithm were employed to solve the models respectively of the two problems mentioned above, and the feasibility and validity of the model and algorithm were validated via illustrative examples.