Dynamics modeling of necessary quantity of relief supplies in large-scale disasters

Accurate demand forecasting is necessary to provide effective support. The dynamics of logistics during disasters are drastically different from normal circumstances. We represent the transition of the amount of goods at each logistics base and evacuation center by a type of discrete time state space modeling. When an earthquake occurs, people evacuate to evacuation centers due to various factors, such as the collapse of houses, stopping of lifelines, and lack of supplies. When these evacuees live in evacuation centers, it is necessary to have a wide range of relief supplies, such as food, drinking water, and household items. Using our proposed model, the necessary quantity of relief supplies for evacuees in evacuation centers can be effectively predicted. We also derive a model to present the dynamics of logistics for relief supply.