A smart model for reducing Gharar in agricultural Salam contracts: using artificial intelligence and geographic information systems in islamic finance

Abstract

The Salam contract represents one of the principal Sharia-compliant financing instruments for agricultural production. However, assessing agricultural land before harvest remains a major source of uncertainty for Islamic financial institutions. Furthermore, existing research generally addresses either the jurisprudential aspects of Salam contracts or the application of Artificial Intelligence and geospatial technologies, without providing an integrated framework that combines legal, economic, and technical dimensions to support financing decisions. This study proposes an integrated decision-support framework combining remote sensing, Geographic Information Systems (GIS), climate data, and Artificial Intelligence to assess agricultural land quality for Salam financing. The empirical analysis is based on 420 spatial observations collected from an agricultural area located in the Casablanca–Settat region of Morocco during the period January 2024 to June 2025. Environmental indicators derived from Sentinel-2 Normalized Difference Vegetation Index (NDVI) imagery and CHIRPS precipitation data were integrated into a geospatial database and analyzed using the K-Means unsupervised clustering algorithm. The results show that K-Means successfully identifies three distinct environmental profiles corresponding to different levels of agricultural land quality. These environmental profiles are subsequently interpreted as financing risk categories, providing an objective basis for Salam financing decisions without relying on predefined classification thresholds or manually assigned labels. The proposed framework therefore contributes to a more transparent and evidence-based assessment of agricultural land while avoiding methodological circularity.

The findings should nevertheless be regarded as preliminary, since they are based on a single study area and a limited observation period. Future research should validate the proposed framework using additional agro-climatic regions, longer time series, and independent agricultural production data in order to assess its robustness and generalizability.

Classification JEL : G23, Q14, O16

Paper type : Empirical Research