Mass Evacuation Transportation Model Using Hybrid Genetic Algorithm

Abstract

The process of evacuating natural disasters requires careful planning. In particular, the evacuation process needs attention in the evacuation process because it involves the safety of many people. Evacuation time itself depends on information about incomplete evacuation routes such as those concerning desired velocity and obstacle parameters. When viewed in terms of transportation planning for evacuation, it is an Auto-Based Evacuation Model problem where the community, in this case, drivers, certainly do not know the evacuation planning or the route they will go through because, in the event of a disaster, it cannot be predicted which areas will be affected. The routing problem can be viewed as a discrete problem where the traffic problem is following a user equilibrium model. It has a bi-level structure. Top-level is used to minimize evacuation time using the contraflow strategy. At the same time, traffic volume and travel time are modeled at a low level. This problem is a linear programming problem whose solution will be optimized using a Hybrid Genetic Algorithm. This model is proposed to carry out mass evacuation processes based on time-window constraints. Finally, computational results are provided to demonstrate the validity and robustness of the proposed model. Based on the test results, it can be seen that the designed model can adjust the path that the vehicle follows with the vehicle station by adjusting the available capacity. The results showed that the intended route provided by the model was the shortest route.