Genetic Algorithm-enhanced Rank aggregation model to measure the performance of Pulp and Paper Industries

Abstract

Performance measurement is a complex but important task required in all sectors. The problem however arises when usage of different methods for performance assessment provides different results. Under such circumstances when there is a difference of opinions, rank aggregation methods can be used to provide the best solution to decision-makers (DMs). Such approaches, also known as data fusion approaches, combine ranked lists from various methods to generate a consensus. In this study, a novel rank aggregation method is proposed for addressing the problem of conflicting MCDM ranking results. The suggested method uses genetic algorithm (GA) to minimize the Euclidean distance between the ideal ranking and the ranking computed by multiple MCDM methods. This model is embedded into a hybrid multi-criteria decision-making (HMCDM) approach, which is divided into three distinct phases. The first phase identifies the most efficient alternatives; the second analyses the rankings obtained through various MCDM methods; and finally, a compromise ranking result is generated. The proposed approach is employed to measure the performance of Indian Pulp and Papermaking Industries (IPPI).