Title: Presenting a predictive benchmark model of after-sales service agencies for vehicles based on the data envelopment analysis approach

Authors: Sajjad Kheyri; Farhad Hosseinzadeh Lotfi; Seyed Esmaeil Najafi; Bijan Rahmani Parchkolaei

Addresses: Department of Industrial Engineering, Islamic Azad University, Central Tehran Branch, Tehran, Iran ' Department of Mathematics, Islamic Azad University, Science and Research Branch, Tehran, Iran ' Department of Industrial Engineering, Islamic Azad University, Science and Research Branch, Tehran, Iran ' Department of Mathematics, Islamic Azad University, Nour Branch, Mazandaran, Iran

Abstract: Everyone is aware of the importance of benchmarking in all industries. The same is true of the automotive industry. One of the methods of continuous improvement of car after-sales service agencies is benchmarking from successful and efficient examples in the country. Given that evaluation and benchmarking methods are usually retrospective, and also, the rapid changes in environment and customer needs, current methods cannot quickly define corrective actions. In this paper, first, a benchmarking model based on data envelopment analysis is developed for car after-sales service dealers as decision-making units, then considering that the model outputs have a high correlation coefficient, an innovative machine learning model has been used to predict the outputs. Finally, the results of the proposed prediction model are compared with a perceptron neural network algorithm. The results show that the benchmarking and prediction model together with a 7.7% error predicts benchmarks for the end of current period.

Keywords: data envelopment analysis; DEA; correlated output prediction; general two-step; genetic algorithm; predictive benchmark.

DOI: 10.1504/IJSOM.2023.133409

International Journal of Services and Operations Management, 2023 Vol.46 No.1, pp.1 - 34

Received: 07 Feb 2021
Accepted: 04 May 2021
Published online: 15 Sep 2023 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article