Multi-Criteria Ship Route Modeling and Optimization Using Ant Colony Optimization with a Case Study in Indonesia

Optimal ship routing is the key to maximizing revenue and minimizing operating costs in the shipping industry. Several factors significantly affect revenue potential, including demand fluctuations at each port, maximum ship capacity limitations, maximum ship voyage time limitations, and capacity limitations of the type of freight carried. In this paper, we develop a model to optimize ship route selection to maximize revenue by considering constraints such as possible port properties, ship capacity, and types of loads carried. This study uses a machine learning-based model called the Ant Colony Optimization (ACO) algorithm to generate routes allowing revenue maximization while meeting various constraints. The case study was conducted on a highly complex shipping route in Indonesian waters, which included limited ship capacity, sailing time constraints, maximum capacity for various types of cargo on each ship, and different fluctuations in cargo demand at each port. As validation, the results obtained from ACO were compared with the results from the Genetic Algorithm (GA) algorithm and the manual route used. The results showed that the optimal route generated by ACO increased revenue by up to 743% compared to the manual route previously used. This demonstrates the significant potential of ACO in helping shipping companies improve operational efficiency and profitability.