SISTEM PENDUKUNG KEPUTUSAN MOBILE KETAHANAN PANGAN BERBASIS ANALITIK HARGA DAN DISTRIBUSI DI KABUPATEN LAMONGAN

Abstract

Regional food security requires an information system that integrates price, stock, distribution, field report, and early-warning data into a unified decision-making workflow. This study develops PanganKu as a mobile decision support system prototype for food security monitoring in Lamongan Regency. The research used a three-iteration prototyping approach covering system architecture design, functional requirement formulation, price-distribution analytics implementation, and system verification. Evaluation was conducted through black-box testing, algorithm unit testing, and calibrated simulation-based price scenarios. The system provides price and stock monitoring, simple price prediction, anomaly detection, geographic distance estimation, and preliminary surplus-deficit distribution recommendations. The results show that all functional test cases produced outputs consistent with expected results. Price scenario evaluation produced a mean absolute percentage error of 1.0% for stable commodities and 14.2% for volatile commodities. These results indicate consistent algorithm behavior in limited scenarios rather than market-level prediction accuracy. Operational implementation still requires official daily data, end-user testing, field validation, and structured load testing.