Weather downtime prediction in a South African port environment

Abstract

Sea ports act as a gateway for a country’s imports and ex-ports. Delays of vessels at the anchorage due to adverse weather eventsare becoming increasingly problematic. This paper investigates the useweather data to accurately predict delays experienced by ships at theport anchorage by means of both regression (delay duration) and clas-sification (delay impact). The datasets used in this paper consist of fiveyears of weather information and vessel weather delay data obtained fora South African port. The weather information consist of three datasources, including rainfall, wind and wave data. An artificial neural net-work (ANN) was found to perform the best in the prediction of vessel weather delay duration for both three day and weekly datasets and a random forest performed the best in predicting likelihood of weekly vessel weather delays.