عنوان مقاله [English]
Aim and Introduction: Today's complex environment confronts organizations with the constant challenge of achieving goals. The existence of a performance appraisal system to monitor the performance of the organization in achieving goals is inevitable. Organizations in today's competitive environment are successful if they put their performance appraisal on the agenda with high accuracy. Therefore, the main purpose of this paper is to provide an integrated model for evaluating the performance of decision-making units with appropriate resolution (Case study: higher education institutions).
Methodology: The research has a descriptive-explanatory aspect in terms of method. In terms of research orientation, it is developmental and has used survey strategies. The statistical population includes university experts in Semnan, 57 of whom were randomly selected. The data collection tool has developed two questionnaires whose content validity and structure are based on expert opinion and factor analysis; Their reliability has been confirmed based on Cronbach's alpha value of more than 0.7. The fuzzy inference rules were also developed based on the opinions of five selected academic experts. "Data Envelopment Analysis" was also used to evaluate the performance of the mathematical planning approach.
Findings: The design of the integrated model began with the identification of appropriate indicators and structures for evaluating the performance of higher education institutions. Findings include the identification of 9 input variables and 17 output variables. After performing factor analysis on the indicators, three structures of student admission status; manpower status, and infrastructure status with input nature and three structures of educational and research status of students; status of university incomes and status of university services were identified as output structures. Another finding of the research is the development of 27 rules for calculating the value of each structure by the Mamdani fuzzy inference system. The results showed that the proposed model evaluated 6 units as efficient, which is much less than the other four models identified as efficient. Also, the results of the Kruskal-Wallis test show that the average efficiency rating of the proposed model is lower than other models, which confirms this claim.
Discussion and conclusion: Organizations, especially in today's complex environment, need performance appraisal to be aware of the success of their activities. Performance appraisal provides the basis for continuous improvement, growth, and development of organizations. Performance evaluation determines the achievement of the organization's goals. Managers are aware of the challenges ahead and determine the success rate of policies implemented. Performance appraisal improves employee motivation in line with the desired behavior of the organization and is one of the requirements of the organization. Due to the necessity of discussing various models with different origins, it has been designed to evaluate performance. Meanwhile, the data envelopment analysis model is one of the most widely used models in the field of operations research due to its advantages. The multiplicity of input and output variables in data envelopment analysis poses the challenge of reducing the resolution of decision-making units. Various methods have been developed to address this shortcoming. In the meantime, reducing input and output variables by considering the preservation of information contained in them is one of the common solutions. Despite many studies, there is still no agreement among researchers on the best way to reduce variables. A review of the research background showed that two objective and subjective approaches are often used separately to reduce the variables. Considering the advantages of both approaches and neglecting their simultaneous use in most studies, the integrated research model is designed based on subjective approaches that focus on the opinions of experts and an objective approach that has tried to reduce variables according to the data model. Is. In this way, the results of performance evaluation are as real as possible and at the same time with high resolution.