A reliable probabilistic risk-based decision-making method: Bayesian Technique for Order of Preference by Similarity to Ideal Solution (B-TOPSIS)

The Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) is a widely accepted and applied tool among all multi-criteria decision-making (MCDM) methods. Conventionally, TOPSIS finds the relative closeness weight to the ideal solution according to the preference of a single decision-maker. In other words, it fails to integrate the preferences obtained from a group of decision-makers in a decision-making problem. Since the first TOPSIS was proposed, many aggregation preference procedures such as geometric or arithmetic mean using multiple decision-makers have been developed. It is clear that the most previously applied methods are overwhelmingly sensitive to the data outliers that subsequently limit information regarding overall preferences obtained from all decision-makers. This study proposes an innovative methodology by developing a Bayesian TOPSIS (B-TOPSIS) model to aggregate the final weight of alternatives for a group of decision-makers. For this purpose, the TOPSIS framework is modified by considering a probabilistic perspective. The hierarchical Bayesian model is presented to obtain the vector weight for the availability of multiple decision-makers. To illustrate the efficiency and feasibility of the introduced B-TOPSIS, selecting a proper corrosion treatment plan in an oil and gas industrial sector as a real case study has been applied. The results obtained from B-TOPSIS were compared with the conventional TOPSIS results. The priority of alternatives is Biocide treatment, Pigging, Combination Biocide treatment and Pigging, and Monitoring and control parameter over time. The outcomes confirm that the proposed model has a significant advantage because it uses much more information than the original form of TOPSIS.