This paper establishes the standardization of an electro-chemical method called wet oxidation potential (WOP) technique for determining the susceptibility of coal to spontaneous combustion. A total of 78 coal samples collected from 13 different mining companies, spread over most of the Indian Coalfields, have been used for this investigation. Experiments were carried out at different concentrations of KMnO4, viz., 0.05, 0.1, 0.15, and 0.2 N in 1 N KOH and at 27, 40, and 45 °C. With a combination of different concentrations of KMnO4 and temperature, 12 experiments were carried out for each coal sample. Altogether, 936 experiments have been carried out by adopting different experimental conditions to standardize WOP method for wider applications in mining industries. Physical, chemical, and petrographical compositions of coal samples were studied by proximate, ultimate, and petrographic analyses. In order to determine the best combinations of experimental conditions to achieve optimum results in wet oxidation potential method, results were first analyzed by principal component analysis and then by artificial neural network analysis. These analyses clearly reveal that susceptibility index “rate of reduction of potential difference” (RPD12), keeping experimental condition with 0.2 N KMnO4 in 1 N KOH solution at 45 °C, produces optimal results in finding out the susceptibility of coal to spontaneous heating. Further, coals are classified according to their proneness to spontaneous heating with multilayer perceptron (MLP) classifier. A correct classification with accuracy of 94.29 % on test data has been achieved with this classifier. The results have been further validated by tenfold cross-validation method to show its consistent performance over the chosen features.