A HYBRID REGRESSION APPROACH OF BLOCKCHAIN-BASED CROP AND YIELD GAP ESTIMATION USING HEURISTIC OPTIMIZATION

Abstract

Farmers are experiencing reduced crop yields and income due to unpredictable weather patterns and climate change, leading to crop losses from extreme conditions, pests, and diseases. These impacts can affect farmers' livelihoods and national food security, particularly in vulnerable regions in the months of January to August, 2025. The paper is on Blockchain Technology and Lasso regression for estimating crop yield gaps, crucial for agriculture's role in food security, economic growth, and environmental stewardship. It addresses agricultural challenges such as climate change and soil degradation, proposing a blockchain-based approach to overcome limitations in current yield gap estimation methods using Neutrosophic logic. The methodology involves a graph theory model for supply chain optimization and a Lasso regression model for yield gap calculations using heuristic optimization. The Kolmogorov-Smirnov Test is used to determine if the actual yield and potential yield are the same. An illustrative example demonstrates a yield gap of 0.6 tons/ha, underscoring how understanding these gaps can improve crop management and productivity. The crop yield estimate values are used to predict damage based on weight and distance for one acre of small farmers to conduct a multi-regression analysis.