<![CDATA[Indonesia is a country that is highly vulnerable to natural disasters, necessitating a rapid and systematic post-disaster rehabilitation and reconstruction process. However, regional disaster management agencies frequently encounter subjective and inconsistent damage assessments from field surveyors due to the absence of clear standardized criteria. To address this issue, this study develops a web-based Intelligence Decision Support System Dynamic (IDSSD). The system integrates a classically modified Multi-Criteria Decision Making (MCDM) method, specifically Modified-TOPSIS, with an Artificial Intelligence (AI) model utilizing a Single Layer Perceptron Neural Network. The Modified-TOPSIS method utilizes Pairwise Comparison to objectively determine criteria weights, while the Neural Network dynamically adjusts criteria weights when encountering new or unpatterned data. The system was evaluated using confusion matrix parameters (accuracy, precision, recall, and F-measure) based on 16 field datasets from BPBD Blitar, East Java. The experimental results demonstrate that the Modified-TOPSIS method achieves an accuracy of 75% (classified as Fair Classification), whereas the integrated Neural Network-Modified TOPSIS method achieves an accuracy of 81% (classified as Best Classification). The integration of Neural Network significantly enhances system adaptability and objectivity in assessing post-disaster damage.]]>
Copyrights © 2026
