The Study Of Last Mile Transportation Problem In Humanitarian Relief Logistics

In recent years, emergencies (nature disasters, violent confrontations, spreading plagues etc) took place frequently. The timely accurate and efficient distribution of emergency supplies from distribution centres to victims, namely the last mile transportation had became the core of emergency response determining whether the disaster relief operation was success or not. Usually, the objective of the last mile transportation problem is the minimization of transportation cost (efficiency objective) or the maximization of a certain service level indicator (efficacy objective, for example the satisfied amount of demand or the speed of distribution). But these objectives always led to inequity allocation of supplies:some areas closer to the distribution centre might receive materials far later than the remote regions and the arriving time of the first batch of materials among disaster areas could be much different. In this situation, the results not only deviate from the originalintention of disaster relief logistics to save more people as soon as possible and might cause the expansion of thedisaster. So how to determine the appropriate objective for the last mile transportation so as to provide the same level of service and a certain percentage of emergency supplies to all victims in the same time as much as possible is becoming the key to reduce the society harm of a disaster. This thesis mainly settles two aspects of the last mile transportation problem:modelling and analysis of the last mile transportation problem with equity objective, and the choice and integration of objectives among efficiency efficacy and equity.1、when a large-scale disaster breaks out, the emergency supplies and the number of available vehicles are usually limited. A split delivery model for emergency supplies is proposed to deal with this situation. The total amount of demands are split and satisfied within multi-periods by multiple vehicles. This thesis defines a disutility function expressed by the unmet demand proportion of victims, and then the equity objective can be calculated by the sum of the disutility function value among all victims within the whole periods, which realizes the equilibrium of the distribution among different victims. The analysis of results indicates that in contrast to the objective of minimization of transportation cost and arriving time, the equity objective can dispatch a high proportion of relief supplies to all victims in the shortest time period.2、Based on the classification of emergency materials and uncertainty of supplies, this thesis constructs a multi-period operational model for the last mile transportation problem with objectives of efficiency and equity. The efficiency objective calculates all vehicle routing costs within the planning horizon and the equity objective balances the punishment costs among all disaster areas caused by the unmet demand proportion of victims. The model is compiled with the Yalmip toolbox of Matlab, the results of numerical examples indicates:Punishment cost parameters can decide the priority level of different emergency supplies, only when the punishment cost parameter of one kind of supplies is greater than the marginal value then this kind of supplies can be distributed, when the punishment cost parameters of different supplies are same, the shipping quantities of one kind of supplies in every period is indirect proportion to the amounts of this kind of supplies demanded in disaster areas, but the unmet demand proportion among different kinds of supplies still keeps consistent.This thesis not only provides some references for the appropriate allocation of emergency supplies and the choice of appropriate objective in the last mile transportation problem, but also helps improving the operation efficiency of disaster relief logistics.