Study On Site Selection Of Disaster Relief Materials Depository Considering Failure Risk

The prior storage of relief supplies is a guarantee for a nation and a region to cope with the nature disasters. Reasonable reserve location can not only improve the capacity of disaster response, but also reduce the cost of disaster relief. In real life, supplies storehouse fails due to many reasons such as poor weather, natural disaster and so on. Meanwhile, because of the urgency and timeliness of disaster relief, once the supplies storehouse fails, it will result in serious loss of life and property. However, in the existing research work, there is almost no investigation in considering the disruptions of supplies storehouse allocation design.To improve the reliability of the relief supplies reserve network, we propose a strategy of backup supplies, that is to say, to prepare extra relief supplies storehouse for the potential demand places in the designing phase of the relief supplies reserve network. For this strategy, we propose a location model of the relief supplies storehouses considering facility disruptions. The objective of the model is to minimize the sum of the location cost and the expected disruption cost, to determine the location, capacity of storehouses and the demand allocation simultaneously. The model is a NP-hard problem typically, considering the complexity of the model, we employ the GA (genetic algorithm) to solve it, and verified the effective of the model and the feasible of the algorithm through 6 numerical examples.The solution of the examples shows GA have significant advantages in solving time and solving scale comparing with the optimization method The errors of GA were deemed within acceptable limits. Comparing with the model we developed with the location model of non-considering disruptions the total economic cost has a certain increase, but with the continuous expansion of the network, the increase of economic cost and the size of the increase will shrink continuously. Finally, we apply the model to Kunming example getting the location, capacity of storehouses and service area; the parameter sensitivity analysis shows that a critical time constraints existed in the system, when surpass the time restriction, location and material repertory scheme is no longer change, and the critical value can be a best delivery time limit that the relief supplies transported to the disaster area.