Hub Location Optimization for Disaster Aid Distribution

Abstract

The distribution of disaster relief needs to be done at a quick and efficient cost. The distribution and cost-efficient are closely related to the determination of the recipient a hub and the distribution model. This study proposes a disaster relief distribution model by considering the position and number of Hubs for various scenarios. The hub position is essential because it will affect the speed and cost of distribution. The existing Hub and the Hub's determination with the clustering method using the K-Means Clustering algorithm try to propose to this study. Hub distribution model Capacitated Vehicle Routing Problem with Time Windows (CVRPTW) is executed by using the vehicle routing problem feature in Geographical Information Systems (GIS). The distribution of rice aid commodities during the Covid-19 disaster was used to test the validity of the model that had been made. From results, it is found that the addition of a hub can reduce vehicle use by up to 28% and speed up the distribution process from 14 days to 13,12 and 11 days with a total cost that the difference is not significant but increases the service level.