Mavi yaka performans değerlendirmesinde kullanılacak ölçütlerin belirlenmesi için AHP temelli bir matematiksel model.

The criteria to be used for evaluating the performance of workers vary much more than those used to evaluate the performance of office personnel. Therefore, it is more difficult to determine which criteria to use. All criteria, as much as possible, should be taken into account to evaluate the performance healthily way. However, each new criterion defined requires the collection and evaluation of a new data set. However, each new data group that needs to be collected for each employee means a new workload for businesses, and therefore business managers want to be able to evaluate the performance of their workers in the healthiest way, using as few criteria as possible. In this study, a mathematical model was proposed to determine which criteria to be used in evaluating worker performance with AHP, and the proposed approach is applied using data from a sample business and the obtained results are discussed. The criteria to be used in classical performance evaluation studies are generally determined by experts only considering their contribution to performance evaluation. In this case, the data collection workload can be too high and it may make it impossible to use the method in practice. In this study, firstly, workers' performances were calculated with Analytic Hierarchy Process using the criteria determined by the experts, and then, using the proposed multi-objective mathematical model, the criteria that could represent all the criteria with the minimum total data collection workload were determined. Since performance evaluation is carried out continuously in certain periods, working with fewer criteria that can represent all criteria ensures that the data collection workload is drastically reduced and thus the proposed method can be used in practical life. The proposed method has been applied in a plastic injection factory. As a result, 30 out of a total of 40 criteria determined by the experts were selected, with a deviation of only 4% from the performance score obtained when all criteria were selected, the workload of obtaining data was reduced by 26%. [ABSTRACT FROM AUTHOR]