Data-driven delivery model for distributing daily necessities to residents in a specific emergency area

A data-driven vehicle transportation model based on the center location and vehicle routes is proposed to delivery daily necessities to communities in a specific area when an emergency happens. The proposed model includes area partition and packages delivery. To solve this model, a two-stage algorithm is designed. In the first stage, the subarea centers are determined according to the K-means clustering. In the second stage, the vehicle routes in each subarea is solved based on the LKH algorithm and used as a feedback to the first stage. The two stages are executed attractively until a feasible optimal solution is obtained. Finally, the empirical analysis is performed based on the data collected from a controlled area in Chongqing city, China in the covid-19 pandemic period. The results show that the proposed model can help government make better plans and timely deliver living materials to communities in a controlled area. Therefore, the proposed model is of great significance to ensure the normal life of residents in the emergency area.