A multitask deep neural network (DNN) is developed to simulate the ozonation of oil-drilling cuttings (ODC) and is interpreted through a technique of explainable artificial intelligence (XAI) to provide knowledge about the experimental conditions that will maximize the decontamination of ODC. On a semi-batch bubble flow column, ozonation experiments of ODC are carried out after pretreatment with synthetic seawater (SW) and the anionic surfactant sodium dodecyl sulphate (SDS). The performance of ozonation experiments is evaluated by measuring the removal efficiency of the total organic carbon (TOC). The experimental data are used for training and testing an DNN that can predict accurately the TOC removal efficiency of the ozonation process as well as the values of different variables such as pH, oxidation-reduction potential (ORP), temperature (T), pressure drop (ΔP), based on the values of the input variables of the model. The acquired model is interpreted through the Shapley Additive explanations (SHAP) method, an important advancement in the field of machine learning interpretation provided by XAI, regarding the significance of the models’ input variables in the TOC removal efficiency. This step aims at establishing the experimental conditions that lead to the highest remediation rate.