Decision Making Modelling on Sustainability Cost for Travel Path

Abstract

The increasing demand for sustainable logistics operations necessitates advanced optimization techniques that balance cost efficiency with environmental responsibility. Existing route optimization models, such as Prim’s and Kruskal’s Minimum Spanning Tree (MST) algorithms, primarily focus on minimizing distance but fail to adapt to real-world constraints such as traffic fluctuations, sudden delivery requests, and fuel price variations. This study introduces an extended autocatalytic algorithm, which improves route adaptability by incorporating dynamic sustainability metrics, including CO2 emissions and fuel consumption. Using a case study of a delivery firm in Johor Bahru operating a Nissan Commercial Cabstar UD 40, the proposed method is compared against traditional MST-based approaches. Results indicate that while the extended algorithm enhances route flexibility, it incurs a 68.7% higher cost and 22.1% higher CO2 emissions due to the enforced return path. However, sensitivity analysis reveals that the extended algorithm is more resilient to fluctuating fuel prices and road congestion, making it suitable for dynamic logistics operations. Compared to traditional MST models, the extended autocatalytic algorithm offers superior adaptability, allowing real-time adjustments to delivery networks at the expense of slightly higher costs. The findings suggest that integrating real-time traffic data and AI-driven decision models could further refine the algorithm, making it more practical for e-hailing services, urban delivery networks, and fleet management systems. This research contributes to sustainable logistics by providing a scalable, data-driven solution that optimizes delivery routes while balancing economic and environmental objectives.