Improved Entropy Weighting Model in Water Quality Evaluation

Entropy weighting model (EWM) is a widely used weighting method in water quality assessment. EWM assigns weights on the basis of the dipartite degree principle. A large weight is assigned for a pollutant with high dipartite degree, and vice versa. However, this dipartite degree principle cannot properly represent the pollutant's importance through frequent practice when its observation data focus on the worst category. Therefore, weight parameters become illogical. In this study, we reveal this problem through a typical example generated via Monte Carlo simulation. Then, the conventional EWM is improved on the basis of relative entropy theory. In comparison with the conventional EWM, the improved EWM can comprehensively represent indicators' dipartite degrees and pollution conditions, thereby increasing the rationality of weight results.