Title: A hybrid approach to evaluate employee performance using MCDA and artificial neural networks

Authors: Malik Haddad; David A. Sanders

Addresses: Northeastern University – London, UK ' School of Mechanical and Design Engineering, University of Portsmouth, Portsmouth, UK

Abstract: A new hybrid approach for employee performance evaluation based on multiple criteria decision analysis (MCDA) and artificial neural network (ANN) is presented. This is the first time this type of ANNs has been used for this application. A deep ANN is created. A MCDA method used randomly generated sets for training and testing the ANN. The network provided 93.63% training accuracy and 91.91% testing accuracy when tested against the training and testing sets respectively. The new approach could be transformed into a generic employee evaluation tool suitable to accommodate any number of employees and evaluation criteria using transfer-learning. A real-life employee evaluation problem is used as an example. Six employees and six evaluation criteria are considered. The new approach successfully identified the employee most eligible for promotion and ranked the other employees according to their performance.

Keywords: artificial neural networks; employee evaluation; employee performance; MCDA; transfer-learning.

DOI: 10.1504/IJMDM.2024.135301

International Journal of Management and Decision Making, 2024 Vol.23 No.1, pp.58 - 76

Received: 28 Mar 2022
Accepted: 29 Aug 2022

Published online: 04 Dec 2023 *

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