A new learning interaction rule for municipal household waste classification behavior based on multi-agent-based simulation

Abstract

The sorting and recycling of municipal household waste (MHW) as a complex adaptive system is becoming increasingly popular in current research. However, traditional simulation modeling ignores the important influence of anticipated regret and personality traits on individual choice behavior. This study introduces the degree of regret-joy in agent-based simulation modeling and designs a learning rule that takes into account the personality characteristics of individuals, their neighbors, and historical information to analyze the waste sorting behavior of municipal residents. The results suggest that the number of people participating in formal recycling approaches is significantly higher than informal recycling approaches after accounting for the economic profits of formal recycling approaches. Our results also reveal that subsidies are an important factor influencing the waste sorting behavior of municipal residents. However, simply reducing the time required for a formal recycling approach does not affect the behavioral decisions of municipal residents. Only improving the accessibility of facilities in formal recycling approaches on the premise of regulating the informal recycling market, will the number of municipal residents participating in the former increase significantly. The intensity of communication and learning among municipal residents also affects the volatility of decision-making. When intensity is high, group decision-making changes greatly, but at low intensity, group behavior tends to be stable. These findings are useful as a theoretical reference for the development of waste management policies.