Artificial Neural Network (ANN)

– based novel Performance evaluation technique



by Ashwani Kharola, Ravindra Mamgain, Ankit Jain

Department of Mechanical Engineering
Tula’s Institute

Dehradun, India



In this study we have highlighted different performance evaluation techniques in order to carry out adequate performance based appraisal of employees working in different organisations. The paper briefly describes and compares both traditional as well as several modern Performance appraisal techniques which are currently in practice. In this study we have proposed a new soft-computing based appraisal technique which works by combining learning ability of artificial neural networks and reasoning capability of fuzzy logic theory. The results are validated by comparing working of proposed approach to conventional average based rating technique. The results further confirmed appropriateness of above mentioned technique over other traditional appraisal approaches.

Keywords: Artificial neural network, Performance appraisal, soft-computing, ANFIS, fuzzy, Matlab.

  1. Introduction to Performance appraisal

Performance appraisal of an employee is a significant factor in success of any individual as well as for the growth of any organisation. An appraisal is considered to be good if it is performed fairly and motivates employees thereby resulting in improved performance of the organisation [1]. Sometimes performance appraisal does not results into a valid and reliable evaluation thus creating conflicts in the workplace [2]. These conflicts will affect the output and performance of both the employees as well as the organisation. Performance appraisal plays a vital role in both human resource management as well as strategic management and therefore widely employed for both theoretical and practical study [3]. In the past few years researchers have been showing keen interest to develop various appraisal tools and techniques [4]. A case study on effects of performance appraisal in the Norwegian municipal health services was carried out by Vasset et al. [5]. In the study authors evaluated the effect of job motivation, learning and self-assessment on performances of health personals. Shaout and Yousif [6] highlighted various methods and techniques for performance evaluation of employees. The study considered both traditional as well as new approaches for effective appraisals. The authors further proposed a fuzzy based appraisal technique for evaluating performances of academic staff in Sudanese Universities.

A new construct for performance evaluation of teachers was proposed by Yonghong and Chongde [7]. The study conducted literature survey, case study, interview and qualitative research for analysing reliability and validity of different empirical approaches. Keaveny and McGann [8] observed the effectiveness of different performance appraisal formats in terms of clarity. The different formats which were adopted for analysis were simple graphic rating scale, complete graphic rating scale and behavioural rating scales. The study highlighted superior performance of behavioured rating scale compared to other two scales. Jawahar [9] demonstrated a correlation between employee satisfaction and their performance feedback. The study considered a survey on 112 employees which proved that satisfaction with appraisal feedback was directly related to job satisfaction and organisational commitment and inversely related to turnover intentions. Banner and Cooke [10] explained some of the main conceptual issues in performance appraisals. The authors highlighted some practical dilemmas and their solutions which may arise during process of appraisals. The results concluded that one can morally justify use of appraisals under certain specific conditions. Osmani and Maliqi [11] carried out a study to examine the process of management and performance evaluation of employees. Authors focused on importance of individual performances, stages through which appraisal is realised, targets, key indicators and challenges faced during the process in both public and private organisations.

Sanyal and Biswas [12] examined the attitude of employees towards performance appraisal in software companies in West Bengal (India). A survey on 506 employees from 19 software companies was carried out followed by binary regression analysis. The authors identified main consequences of performance appraisal and their impact on motivation of employees. Min-peng et al. [13] proposed a fuzzy comprehensive evaluation approach for measuring performance of engineering R&D staff. The study considered different performance indicators based on morality, ability, diligence and performance to determine weight of every index. The proposed model is feasible and practical through empirical research. Wu and Hou [14] proposed an employee performance estimation model for logistics industry. The proposed model includes three modules for performance estimation i.e. direct performance determination, indirect performance determination and performance score analysis. The proposed model helped in accurate estimation of employee performance. Katerina et al. [15] identified different performance appraisal methods in agricultural organisations. The study initially described some formal appraisal techniques and further designed a questionnaire to rate different appraisal techniques in agriculture sector of Czech Republic. The results showed that most widely used techniques for performance appraisal includes goal-based appraisal, predefined standard outcome-based appraisal and appraisal interviews.


To read entire paper, click here


About the Authors

Ashwani Kharola

Tula’s Institute
Dehradun, India


Mr. Ashwani Kharola
received B.Tech (with Honors) in Mechanical Engineering from Dehradun Institute of Technology, Dehradun in 2010 and M.Tech in CAD/CAM & Robotics from Graphic Era University, Dehradun in 2013. Presently he is working as Assistant Professor in Department of Mechanical Engineering at Tula’s Institute, Dehradun. Earlier he has worked as a Research Fellow in Institute of Technology Management (ITM), One of premier training institute of Defence Research & Development Organisation (DRDO), Ministry of Defence, Govt. of India. He is pursuing PhD in Mechanical Engineering from Graphic Era University (Deemed University), Dehradun. He has published many Research papers in National/International peer reviewed ISSN Journals and IEEE Conferences. His current areas of work includes Fuzzy logic reasoning, Adaptive Neuro-fuzzy inference system (ANFIS) control, Neural Networks, PID, Mathematical Modeling & Simulation. He can be contacted at [email protected]


Ravindra Mamgain

Tula’s Institute
Dehradun, India


Mr. Ravindra Mamgain
completed his graduation in Mechanical Engineering from CCS University Meerut. He completed his Master’s degree in Mechanical Engineering from SLIET, Longowal University. Presently he is pursuing PhD from Uttarakhand Technical University, Dehradun. He has a total 12 years of Teaching experience and received best faculty award in 2008. He has also Published two books under his name. Presently he is working as Assistant Professor and HOD in Department of Mechanical Engineering at Tula’s Institute, Dehradun. He can be contacted at [email protected]


Ankit Jain

Tula’s Institute
Dehradun, India


Mr. Ankit Jain
has completed his graduation in Industrial & Production Engineering from Dehradun Institute of Technology (DIT), Dehradun in 2009. He has also completed his Master’s degree as M. Tech in Thermal Engineering from DIT. Presently he is pursuing PhD in Mechanical Engineering from Uttarakhand Technical University. He has a total of 8 years of Teaching experience. Presently he is working as Assistant Professor in Department of Mechanical Engineering at Tula’s Institute, Dehradun. He can be contacted at [email protected]