Freight transportation systems contribute significantly to operational inefficiencies and greenhouse gas emissions due to suboptimal routing and poor truck capacity utilization. Traditional logistics planning approaches primarily focus on minimizing distance without incorporating dynamic traffic conditions, fuel efficiency, and environmental constraints. This paper proposes LOADMIND, an Artificial Intelligence (AI)-driven platform designed to enhance truck utilization and reduce emissions through intelligent multi-objective route optimization. The system integrates real-time traffic prediction using machine learning models with a Genetic Algorithm-based optimization engine to determine fuel-efficient and emission-aware routes. A mathematical formulation incorporating distance, fuel consumption, and emission parameters is developed. Experimental evaluation using simulated logistics datasets demonstrates improvements of 18% in truck utilization, 15% reduction in fuel consumption, and 17% reduction in CO₂ emissions compared to conventional shortest-path routing. The results validate the effectiveness of AI-driven optimization for sustainable freight transportation.
Charan et al. (Thu,) studied this question.