Voyage Optimization Algorithm for Intelligent Shipping – Considering Energy Efficiency and Collision Avoidance
Doctoral thesis, 2025
Environmental emissions from shipping pose significant challenges caused by the rapid increase in energy consumption. Voyage optimization system is an valuable tool to address this challenge by enhancing energy efficiency, with optimization algorithms serving as its core, enabling better decision-making. The main objectives of this thesis are to develop voyage optimization algorithms to improve energy efficiency and investigate the capability of voyage optimization algorithms for ship collision avoidance. By achieving these goals, it aims to support intelligent shipping, characterized by enhanced decision-making capabilities. Weather routing, i.e., voyage optimization with the aim to increase energy efficiency in ship operations, rely on ship performance models to estimate energy costs and optimization algorithms to find optimal voyages. However, ship performance models may contain large uncertainties in estimating a ship’s energy consumption and emissions. In addition, optimization algorithms should also consider uncertain and dynamic factors, e.g., weather conditions and market fluctuations, to ensure optimal operations.
To achieve the overall objectives, this thesis first conducts a systematical literature review to help researchers and practitioners clearly understand weather routing and identify opportunities in current research for the development of its optimization algorithms. Based on the review, this thesis proposes two innovative approaches to achieve energy-efficient weather routing, an Isochrone-based predictive optimization algorithm (IPO) and a learning-based multi-objective evolutionary algorithm (L-MOEA). They can effectively minimize fuel consumption and optimize energy efficiency, with the aid of emerging machine learning (ML) techniques. In addition, IPO can be conducted in real-time to address uncertainties in weather routing while considering arrival time, and L-MOEA can consider the essential operational uncertainty due to weather forecast. Furthermore, to ensure reliable operations in practice, this thesis investigates the uncertainty of fuel consumption caused by Specific Fuel Oil Consumption (SFOC) in ship performance models, and the impact of this uncertainty on weather routing. Finally, this thesis extends the research outcome on Isochrone-based algorithms to assist shipping in confined waterways. It seeks to achieve real-time voyage optimization for collision avoidance problems while considering the arrival time, assist on-time transport, and ensure ship operational safety.
It can be concluded that the proposed IPO method can achieve an average of 5% energy savings for weather routing, comparable with L-MOEA. It has also been found that the uncertainty due to ship energy performance models should be carefully considered in decision-making to ensure reliable voyage planning. In addition, the proposed IPO-based collision avoidance algorithm can effectively optimize the voyage in real-time to ensure a ship’s operational safety and on-time arrival, complying with COLREGs in both confined waterways and open waters.
To achieve the overall objectives, this thesis first conducts a systematical literature review to help researchers and practitioners clearly understand weather routing and identify opportunities in current research for the development of its optimization algorithms. Based on the review, this thesis proposes two innovative approaches to achieve energy-efficient weather routing, an Isochrone-based predictive optimization algorithm (IPO) and a learning-based multi-objective evolutionary algorithm (L-MOEA). They can effectively minimize fuel consumption and optimize energy efficiency, with the aid of emerging machine learning (ML) techniques. In addition, IPO can be conducted in real-time to address uncertainties in weather routing while considering arrival time, and L-MOEA can consider the essential operational uncertainty due to weather forecast. Furthermore, to ensure reliable operations in practice, this thesis investigates the uncertainty of fuel consumption caused by Specific Fuel Oil Consumption (SFOC) in ship performance models, and the impact of this uncertainty on weather routing. Finally, this thesis extends the research outcome on Isochrone-based algorithms to assist shipping in confined waterways. It seeks to achieve real-time voyage optimization for collision avoidance problems while considering the arrival time, assist on-time transport, and ensure ship operational safety.
It can be concluded that the proposed IPO method can achieve an average of 5% energy savings for weather routing, comparable with L-MOEA. It has also been found that the uncertainty due to ship energy performance models should be carefully considered in decision-making to ensure reliable voyage planning. In addition, the proposed IPO-based collision avoidance algorithm can effectively optimize the voyage in real-time to ensure a ship’s operational safety and on-time arrival, complying with COLREGs in both confined waterways and open waters.
Collision avoidance
machine learning
estimated time of arrival (ETA)
Isochrone algorithm
energy efficiency
voyage optimization
weather routing
Lecture Hall HA2, Hörsalsvägen 4
Opponent: Jin Wang. Professor of Marine Technology, Liverpool John Moores University, United Kingdom.
Author
Yuhan Chen
Chalmers, Mechanics and Maritime Sciences (M2), Marine Technology
State-of-the-art optimization algorithms in weather routing — ship decision support systems: challenge, taxonomy, and review.
Strategies to improve the isochrone algorithm for ship voyage optimisation
Ships and Offshore Structures,;Vol. 19(2024)p. 2137-2149
Journal article
An Isochrone-Based Predictive Optimization for Efficient Ship Voyage Planning and Execution
IEEE Transactions on Intelligent Transportation Systems,;Vol. 25(2024)p. 18078-18092
Journal article
Learning-based Pareto-optimum routing of ships incorporating uncertain meteorological and oceanographic forecasts
Transportation Research Part E: Logistics and Transportation Review,;Vol. 192(2024)
Journal article
A machine learning method to model stochastic SFOC induced fuel consumption for ship voyage optimization.
Isochrone-based real-time ship collision avoidance complying with COLREGs in confined waterways.
International shipping has long been recognized as an economical and energy-efficient means of transportation. It facilitates over 80% of global merchandise trade by volume with 3% of greenhouse gas (GHG) emissions. However, over the past decade, rapid growth in the shipping market has led to a substantial 20% rise in GHG emissions, presenting serious environmental challenges. In response, the IMO has updated its GHG strategy, introducing a phased approach to decarbonization and urging the shipping industry to take proactive measures to reduce its carbon footprint. To meet these challenges and the IMO’s decarbonization goals, voyage optimization systems become essential to modern shipping operations and widely adopted across the industry. Optimization algorithms form the backbone of these systems, playing a crucial role in enhancing shipping efficiency and sustainability.
In this thesis, voyage optimization algorithms are investigated and developed to achieve intelligent and sustainable maritime transportation, with a focus on improving energy efficiency in weather routing and ensuring ship safety in collision avoidance. The proposed algorithm generates practical voyage plans, achieving an average 5% fuel reduction across various sailing conditions. Its runtime averages 20 seconds, including ship performance predictions, making it suitable for real-time applications. Additionally, uncertainties in weather and fuel consumption are analyzed to enhance the reliability of voyage optimization. The study also explores real-time collision avoidance, incorporating terrain adaptability and regulatory compliance to recommend optimized voyages that ensure both short-distance sailing and ship safety.
In this thesis, voyage optimization algorithms are investigated and developed to achieve intelligent and sustainable maritime transportation, with a focus on improving energy efficiency in weather routing and ensuring ship safety in collision avoidance. The proposed algorithm generates practical voyage plans, achieving an average 5% fuel reduction across various sailing conditions. Its runtime averages 20 seconds, including ship performance predictions, making it suitable for real-time applications. Additionally, uncertainties in weather and fuel consumption are analyzed to enhance the reliability of voyage optimization. The study also explores real-time collision avoidance, incorporating terrain adaptability and regulatory compliance to recommend optimized voyages that ensure both short-distance sailing and ship safety.
AUTOBarge - European training and research network on Autonomous Barges for Smart Inland Shipping
European Commission (EC) (EC/H2020/955768), 2021-10-01 -- 2025-09-30.
Areas of Advance
Transport
Subject Categories (SSIF 2025)
Marine Engineering
Energy Engineering
ISBN
978-91-8103-165-2
Doktorsavhandlingar vid Chalmers tekniska högskola. Ny serie: 5623
Publisher
Chalmers
Lecture Hall HA2, Hörsalsvägen 4
Opponent: Jin Wang. Professor of Marine Technology, Liverpool John Moores University, United Kingdom.