Promoting Urban Regeneration Through Multi-Agent Strategic Interaction Behavior: A Dynamic Decision Model for Industrial Park Renewal

Urban regeneration is critical for addressing contemporary urban challenges, yet its complexity arises from the dynamic interactions among different participants’ preference and strategic behavior factors, making it a multi-agent system driven by strategic behaviors. This study, based on a Chinese urban regeneration case, develops a dynamic evolutionary game model for industrial park renewal to explore the strategic interactions among three key stakeholders: government, social capital, and property owners. The findings reveal three insights: Firstly, the probabilities of social capital participation and property owner cooperation exhibit opposing trends, highlighting conflicting incentives. Secondly, social capital participation follows an inverted U-shaped trajectory with investment ratios, reflecting a strategic trade-off between risk and control; further robustness checks incorporating time delays and phased investments confirm that the curvature of this trajectory is highly sensitive to the project’s development cycle. Thirdly, lower land repayment costs, higher rental income, greater project returns, and a higher profit-sharing ratio promote cooperative strategies among property owners, though this effect remains marginal. The study further demonstrates that non-cooperative behavior among property owners results in a single evolutionary stable strategy (1, 1, 0) where the government repurchases land property rights, and social capital acquires these rights for redevelopment. The findings suggest that this conclusion applies specifically to industrial park renewal in urban centers held by property owners in cities, where it is government-led facilitation, with property owners exiting and social capital entering simultaneously, thereby ensuring alignment of multi-agent strategic behavior in China.