[1M4-CC] Late Breaking Research Session - A

[1M4-CC-02] Toward Simulation Analysis of Subsidy Payment Policies Amid COVID-19 Pandemic: Multi-Objective Optimization in Agent-Based Simulation

Yusuke Goto¹,Hisashi Ichikawa²,Hiroyuki Morita³,Yasuyuki Shirai⁴ (1. Shibaura Institute of Technology, 2. Iwate Prefectural University, 3. Osaka Prefecture University, 4. Daito Bunka University)

The COVID-19 pandemic has had a severe economic impact in Japan now, but the extent of the impact has varied dramatically by industry, employment status, region, and household composition. The government and local governments are attempting to provide people financial support through subsidy payment. For adequate support, the amount of money specifically provided to each household by the various supports needs to be optimized. The purpose of the subsidy payment can be defined from various concerning viewpoints, and then we face a multi-objective optimization problem. Thus, a simulation platform is necessary for designing subsidy payment policies by governments and local governments.
This study discusses a simulation model for analyzing the problems mentioned above and issues in the analysis. The model simulates changes in the economic situation of each household based on economic shock scenarios. The current income of each person increases or decreases in response to an economic impact scenario. As a consequence of the economic impact, individuals may become unemployed or employed. The target region is virtually reproduced using synthetic population data in the model, and real-scale simulation is performed.
We recognize that the following issues in simulation analysis have been addressed.
1) Considering the existence of uncertainty in the simulation.
2) Developing an algorithm to obtain the Pareto front in multi-objective optimization efficiently.
3) Considering quality indicators of solution sets for future political decision-making.