Optimization method for territorial spatial planning zoning in Fangchenggang City based on improved simulated annealing

With the gradual transformation of China’s economy from the traditional quantitative economy to high-quality economy, the problems of urban sprawl and land use conflicts left over from the rapid development of industrialization need to be solved. The rational utilization of land resources holds significant importance for the sustainable development of economy and society. Land spatial planning and zoning are increasingly becoming core means to guide the rational utilization of land, implement policy regulation, and promote the modernization of the national governance system. To address the limitations of existing comprehensive index evaluation methods in terms of scientificity and practicality in planning and zoning delineation, a planning and zoning method based on improved simulated annealing is proposed. Based on land use data from 2010 to 2020, the FLUS-Markov model was used to predict the land use status of Fangchenggang City in 2030. Subsequently, the method constructed the future land use pattern of the region based on an advantage weighting approach. Finally, based on the improved simulated annealing method, the optimal planning and zoning scheme were determined through large-scale iteration and an improved Calinski Harabasz index based on land type attribute differences. An empirical study was conducted using Fangchenggang City in Guangxi as an example. The main research findings are as follows: 1) The constructed land use pattern is more aligned with real-world physical boundaries and meets the needs of planning and zoning compared to the results of the FLUS-Markov model; 2) Introducing an improved simulated annealing algorithm with iterative functionality that balances global and local optimization into the planning and zoning method is feasible and advantageous; 3) The scientificity and efficiency of planning and zoning are improved through the following improvements to the simulated annealing algorithm: generating an initial scheme based on ecological corridors and land type agglomeration areas, and improving convergence speed through iterative zoning schemes with memory; achieving optimal zoning scheme selection by balancing zoning accuracy and practicality; and further enhancing the scientificity of planning and zoning results by using zoning evaluation indicators based on land type attribute differences. Compared to traditional zoning methods that can only generate a small number of planning schemes, the method proposed in this paper can generate a large number of planning schemes, and the results can be directly applied to policy implementation, solving the problem of strong dependence of planning zoning on subjective input from decision makers, and improving the scientific and practical nature of planning zoning.