Artificial Intelligence for Geospatial Decision Support in Rural Wildfire Management: A Configurational Mapping Review

Wildfires are increasingly complex and geographically dynamic phenomena that require timely and context-sensitive decision support across the management cycle. Artificial intelligence (AI) has been widely applied to wildfire detection, prediction, and remote sensing; however, a systemic understanding of how AI methods are structurally integrated into decision-support architectures remains limited. The present configurational mapping review, reported in alignment with PRISMA-ScR guidance, examines AI applications in rural wildfire management between 2020 and 2024. Using a configurational framework, explicit scope–algorithm–vector relations are mapped, identifying how specific AI paradigms are operationalised through technological infrastructures to support decision-relevant functions. A total of 27 articles were included, from which 168 scope–algorithm–vector triplets were extracted and analysed descriptively. The results reveal a concentration of applications in detection and evolution prediction tasks, predominantly supported by machine learning methods and remote sensing platforms. Explicitly linked configurations to action-oriented or prescriptive decision functions are less frequently documented. The findings contribute to a structured mapping of AI deployment patterns in wildfire management and provide a conceptual basis for future research addressing integrative and action-oriented system design.