Using the Dual Concept of Evolutionary Game and Reinforcement Learning in Support of Decision-Making Process of Community Regeneration—Case Study in Shanghai

Abstract

Under the digital revolution that spawned in recent years, AI support is raised in the context of urban design and governance as it aims to match the operation of the urban developing process. It offers more chances for ensuring equality in public participation and empowerment, with the possibility of projection and computation of integrated social, cultural, and physical spaces. Therefore, this research explored how scenario simulation of social attributes and social interaction dimensions can be incorporated into digital twin city research and development, which is seen as a problem to be addressed in the refinement and planning of future digital platforms and management in terms of decision-making. To achieve the research aim, this paper examined the evolution of social governance state and strain decision models, built a simulation method for the evolution of complex systems of social governance driven by the fusion of data and knowledge, and proposed a system response to residents’ ubiquitous perception and ubiquitous participation. The findings can help inspire the application of computational decision-making support in urban governance, and enhance the internal drive for comprehensive and sustainable urban regeneration. Moreover, they imply the role of the updated iterations of physical space and social interaction on social attributes.

Keywords

• AI technology-driven community regeneration
• decision-making
• evolutionary game
• reinforcement learning