Decision-supporting models for human-reliability based safety promotion in offshore Liquid Natural Gas terminal

Offshore LNG terminal is a place which contains many high-risk operations. Although many protection instruments have been installed to guarantee the safety in offshore LNG terminal, many accidents still happen. Through the investigation of the accidents, people gradually realise that humans dominate the operational safety in offshore LNG terminal. Therefore, people need a human-reliability based decision-making model to make better plans for avoiding human errors. However, there is limited effort for reaching this requirement, so the particular research should be conducted. This study aims to separately introduce some learning algorithms such as neural networks and K-Nearest Neighbours (KNN) into the construction of the human-reliability based decision-supporting models. All these decision-supporting models will be based on the reality in the Beihai Offshore LNG Terminal. Then, the most robust model will be decided by the Mean Square Error (MSE) and the Average Error (AE) of each model.