Multi-objective spatial optimization of forest fire monitoring networks: An integrated GIS-MCDM framework enhanced by improved genetic algorithms

As one of the most destructive and rapidly spreading natural hazards, forest fires pose a severe threat to the stability of ecosystems. To effectively mitigate fire risks, this study proposes a site-selection model that integrates Multi-Criteria Decision Making (MCDM), Genetic Algorithm (GA), and Geographic Information System (GIS), with the aim of optimizing the spatial distribution of forest fire monitoring points and enhancing fire surveillance efficiency. The model is designed with three primary objectives: maximizing monitoring coverage, minimizing road network distance, and optimizing economic costs. To achieve adaptive decision-making, the Analytic Hierarchy Process (AHP) is employed to dynamically allocate objective weights. Building upon this, differential evolution operators and adaptive mechanisms are incorporated to strengthen the GA’s global search capability and convergence performance. Furthermore, GIS combined with the FUCOM method is utilized for suitability analysis of potential monitoring points, effectively excluding restricted zones such as lakes and farmland to ensure the rationality of site allocation. A case study conducted in a high fire-risk region of Shanxi Province, China, demonstrates that the improved GA exhibits superior performance in terms of convergence speed, solution quality, and stability. Moreover, the model enables flexible adjustment of objective weights according to decision-makers’ preferences, thereby generating multiple optimized site-selection schemes. Compared with conventional layouts, the optimized configuration achieves an 18.6% increase in monitoring coverage, along with reductions of 50% in point-to-road distance and 10.2% in economic costs. These findings highlight the effectiveness of the proposed model in multi-objective site selection optimization and provide robust, scientific decision support for the spatial planning of forest fire monitoring networks.