Fuzzy model for predicting contamination of the geological environment during an accidental oil spill

Abstract

Oil spills on the ground cause significant damage to the geological environment, including groundwater, contaminating it and making it uninhabitable. Therefore, it is necessary to predict the scale of the oil spill and the consequences of contamination of the geological environment to minimize the damage. In the literature, there are plenty of approaches for oil spill prediction on the water. However, there is a lack of approaches for predicting oil spills in the geological environment. This paper proposes a new fuzzy model for predicting oil spills on the ground. It uses fuzzy set theory to express the uncertainty of the spilled oil volume and the specific oil capacity to make predictions more efficient and effective. It consists of the following parts: formulating a hypothesis based on initial oil spill data, fuzzy modelling of oil spill penetration, evaluating the hypothesis of whether the spilled oil will penetrate the ground layer and groundwater and finally evaluating the effectiveness of the proposition. The accuracy and reliability of the proposed model were assessed using synthetic data of oil spill penetrations into the ground and predictions of nine experts. The obtained experimental results show that the proposed fuzzy model is valid and does not contradict reality. Furthermore, statistical parameter (MAE and RMSE) shows that the proposed fuzzy model can predict the geological oil spill consequences with sufficient accuracy. It is practical and contributes to the body of knowledge in predicting geological oil spills. In addition, it will assist the practitioner in making decisions about how to respond to an oil spill.