Statistical modeling of bilge water discharge from ships during normal operation

Abstract

This study estimated the amount of bilge water spilled from ships during normal operation and identified the contributing factors of the discharge by building a statistical model. To build the statistical model, we collected as much information as possible about the amount of bilge water in ships, then used the collected information to formulate a probability distribution of the discharge amount according to Bayesian statistics, and determined the parameters included in the model by applying a Markov Chain Monte Carlo method. The analysis of those parameters shows that the integrated bilge treatment system (IBTS) primarily contributes to reductions in the discharge amount, and that the container-type ship is involved with especially large discharge amounts.