Enhancing Operational Efficiency in Marine Shipping: A Data-Driven Approach to Fleet Performance

Abstract

Marine shipping has a vital role in international trade with operational efficiency being a factor of great concern in terms of cost reduction, reliability, and sustenance of maritime operations. This paper explores how the use of data-based solutions could improve the performance of fleets by examining the main variables of operation, such as fuel consumption, vessel speed, time of voyage, travelled distance, cargo utilization and delays. To identify the patterns of performance and relationships, a quantitative approach was chosen and secondary operational data were used where the descriptive statistics, correlation analysis and regression analysis with the help of machine-learning were conducted. The findings showed that there was a strong positive correlation between the speed of vessels and fuel consumption and distance travelled had significant association with the fuel requirements. The regression analysis revealed that the speed of the vessel, distance, cargo utilisation, and voyage period were significant variables that explained the fuel consumption. Predictive modelling also showed the capability of machine-learning capabilities to predict fleets. The R2 of the Gradient Boosting model was 0.87, which means that this model has strong predictive power in the analysed data structure. The results show that evidence-based fleet management, better voyage planning with statistical analysis and the ability to enhance vessel utilisation and find opportunities to decrease unnecessary fuel consumption could be facilitated with the help of integrating statistical analysis with predictive modelling. The researchers conclude that it is possible to use data-based analysis of fleet performance to make marine shipping operations more efficient and sustainable.