Comparison of AHP, PAPRICA, PROMETHEE, DEX and TOPSIS on an Application for Employee Selection

摘要

Employee selection is an essential process that in many organizations depends only on human judgement, and in today's fast-changing work environment, is susceptible to human errors. Hence, the problem of selecting the most appropriate employee is complex and costly, especially if the selected employee is unsuitable for the job position. The complexity of the problem arises from several requirements: each job position requires more than one criterion to be fulfilled by the job candidate; each candidate has a different set of skills, and usually, several candidates apply to one job position. The decision-maker has to make a quick selection decision, as the longer employee selection process is, the greater the costs are. In this article, we build 20 decision support models for four different job positions with five Multi-Criteria Decision Making (MCDM) methods and we compare them on a real set of data from an employment agency. The goal of this comparison is to recommend which method is most suitable by comparing the correctness of the results, ranking with missing values and difficulty to use. The results show which MCDM method is better for filtering most suitable employees given all required criteria and which MCDM method would be recommendable for employees ranking.