Clustering and Data Envelopment Analysis Integration: Performance Evaluation of Textile Companies in ISO 500

Abstract

Purpose- In a rapidly globalizing world, it is critical to ensure that businesses work effectively and efficiently. The way to work effectively and efficiently is possible by making continuous performance evaluations and making improvements in line with the determined targets. The aim of this study is to measure the efficiency of textile companies in the Istanbul Chamber of Industry (ISO) 500 list in performance management. Design/methodology/approach- The efficiency of the companies selected for analysis in performance management were analyzed with the help of both input-oriented and output-oriented Data Envelopment Analysis models, under the assumption of variable returns to scale. As a result of the literature review, three input variables (number of employees, equity, total assets) and 3 output variables (sales from production, EBITDA, exports) were determined. After the data envelopment analysis, cluster analysis was performed using k-means algorithm. Findings- After clustering companies with similar characteristics, effective and ineffective companies were determined with the help of Data Envelopment Analysis. In order to increase their efficiency, suggestions were made to the inactive firms about which input and output variables they should improve. Discussion- Businesses continuously want to increase their performance by trying to achieve the maximum output with the least input component. In this study, the activities of the enterprises operating in the textile sector were determined by determining the input and output variables in line with the literature. The analysis helps businesses in which factors they can increase their performance by making improvements.