Analisis peramalan data kekeringan lahan pertanian di kabupaten lamongan menggunakan metode Fuzzy Time Series-Markov Chain

Abstract

The agricultural sector in Lamongan Regency is severely affected by the threat of drought, potentially affecting food production and farmers' welfare. This study aims to forecast the area of agricultural drought with a very severe category in Lamongan Regency using the Fuzzy Time Series-Markov Chain (FTS-MC) method. The data used includes the average drought area based on the processing of Landsat 8 imagery from 2020–2024, rainfall data, and historical drought data from the Regional Disaster Management Agency. The selection of the FTS-MC method is based on its strength in dealing with uncertain data and modeling drought condition transitions probabilistically. The forecasting results show that the pattern produced by FTS-MC closely follows the actual pattern, with good accuracy indicated by a MAPE value of 13.25%. The forecast result for January 2025 incerased 925,421.43 m². This finding is expected to serve as a reference for local governments and other stakeholders in planning drought mitigation, irrigation management, and more adaptive, data-driven food security strategies.