<![CDATA[The escalating frequency and intensified magnitude of flood disasters, exemplified by the catastrophic events in Aceh, Indonesia, between 2023 and early 2026, have exposed critical vulnerabilities in traditional, reactive emergency management systems. Current frameworks often suffer from information blackouts, misaligned resource allocation, and a “one-size-fits-all” approach that neglects vulnerable populations. This study proposes SIAGA, an innovative mobile-based information system developed through a design science research (DSR) paradigm to optimize post-flood recovery. The methodological framework of this study is built upon a case study of Aceh, Indonesia. The system integrates a sophisticated multi-layered architecture: a machine learning (ML) engine adapted for automated disaster scale determination with explainable AI (XAI) for decision transparency, a hybrid multi-criteria decision making (MCDM) layer utilizing a tripartite weight synthesis and technique for order preference by similarity to the ideal solution (TOPSIS) logic for optimized logistics, and a resilient mesh network protocol for decentralized communication in infrastructure-constrained environments. Requirement engineering, conducted via a survey with 31 flood-affected respondents, validated the critical need for automated status determination and optimized supply chains. Results demonstrate that the SIAGA artifact successfully bridges the gap between complex predictive intelligence and tactical field response. By manifesting a high-fidelity interactive prototype, this research provides a scalable, socio-technical solution that enhances community resilience and ensures mathematically optimized resource distribution during humanitarian crises. ]]>
