Overview and Assessment of Ship Emissions Estimation Models

Abstract This review provides a comprehensive comparative assessment of ship emissions estimation models across the various ship emissions estimation modelling approaches, with a focus on key approaches: top-down, bottom-up, hybrid approaches, and emerging machine learning and artificial intelligence-based techniques as an innovative approach, offering new perspectives and solutions for the maritime industry through its powerful data analysis and pattern recognition capabilities. It delves into important ship emissions estimation models that use advanced data-driven techniques, ranging from simplified to complex, which are continually evolving to improve accuracy by incorporating real-time data, technological advancements, and the impact of environmental policies. Afterwards, this review provides an overall analysis of the ship emissions models, highlighting current research gaps in data availability and quality, challenges in selecting ship emissions estimation methods, and future research topics. Finally, it concludes with future directions in advanced machine learning methods, digital twins, advanced sensor integration, and global monitoring for precise forecasting and regulatory support, all of which are crucial for decarbonization targets.