Decision-support system for the management of truck stays at seaports.

This work presents a decision-support system for predictive analysis of truck arrivals, dwell times and possible congestion situations in seaport, through a methodology that encompasses machine learning algorithms, data visualization tools and discrete-event simulation models. This decision-support system makes it possible to have a detailed planning of the arrival and transit times of trucks in the port to predict land congestion events; and to know in advance the needs in relation to services or stopovers, and how long each truck will be in the port facilities. The system also allows for the planning and sizing of waiting areas according to expected arrivals and dwell times of trucks in port, at the same time that it optimizes port value spaces, and reduces unnecessary journeys and associated emissions, by evaluating the relationship between the sea side and the demand in land accesses. This will eventually help to improve coexistence and port-city integration between urban and heavy traffic. The system has been validated in the Port of Santander (Spain) with satisfactory results in terms of operational improvement. [ABSTRACT FROM AUTHOR]