Vessel speed analytics using satellite-based ship position data

In the maritime supply chain, sailing speed is a key variable governing ship-to-air emissions and the profitability of vessels. While there can be substantial economic gains to be had from dynamic speed optimization, real-life constraints such as weather conditions, contractual limitations and supply chain considerations may govern the speed choice in practice. The increasing availability of high-frequency ship positions and vessel speeds reported by the terrestrial and satellite-based Automated Identification System (AIS) enables formal empirical testing of the determinants of vessel speed. We use a panel data set of nearly 62,000 weekly average speed observations for a fleet of 607 Very Large Crude oil Carriers (VLCCs) during 2013 - 2015 to estimate a multiple regression model with technical, operational and macro-economic variables. We find that macroeconomic variables have a marginal influence only on ballast speeds, while laden speeds are influenced mainly by whether the operator is also the cargo owner.