Entropy-Based Evaluation of Disaster Information Quality in ICS Using Drone Observation

Abstract

Effective disaster response within the Incident Command System (ICS) depends on rapid and diverse situational information. This study quantitatively compared disaster information obtained by a Human observation group and a Drone observation group using entropy-based metrics and an integrated information quality (IQ) score. A simulated disaster field experiment was conducted in Indonesia. The Human observation group (n=20) recorded events through on-site patrol reports, while the Drone observation group (10 flights) recorded events based on aerial observation. Detected disaster events were categorized and analyzed using Shannon entropy, entropy weight method (EWM), and a time-normalized IQ score. Group differences were tested using Welch’s t-test. The Drone observation group detected more disaster events across all major categories and showed higher entropy values, indicating greater information diversity. The mean IQ score was significantly higher in the Drone observation group (10.6) than in the Human observation group (2.5) (p<0.01, large effect size), indicating substantially greater operational information efficiency under aerial observation.