Modelling of Key Performance Indicators for Staff Advancement in Higher Institutions of Learning using Fuzzy Logic Technique

ABSTRACT 

Performance appraisal is a formal management system that provide for the evaluation of the qualities o1 an indjviduals performance in an organization. One of the viable means of motivating staff in higher institutions of learning is to ensure that staff are rewarded with promotion and other benefits as at when due. in practice, this is not automatic as staff advancement is usually based on a number of criteria or indicators that should be aggregated to achieve a fair and transparent judgment. However, the judgment is subjective and not transparent in many higher institutions of learning. The objective of this study, therefore, is to model some key performance indices required to achieve staff advancement in a typical higher institution of learning using a transparent approach. The concept of fuzzy logic techniques being a knowledge representation approach is used in this study, and in the process, the required attributes are identified, transformed and modeled. Specifically, the method follows the following procedures: fuzzifications, application of the fuzzy operators. rule generation, aggregation of the rule output and defuzzification. The implementation is carried out in Matlab software environment and in the process a number of rules were t~enerated, The study shows a standard way of representing staffs achievements to pave way loi’ advancement by following procedures that is free of subjectivity, This study further illustrates graphically the surface view of the relationship that exists among the indicators and the output (decisions); the surface view unveils a resulting output that is directly proportional. By using an interactive interface, it is recommended that the rules generated and other models represented in this study should be developed to a system using any suitable high level language. As an extension of this work, the model developed can achieve a learning capability if the technique of neural network is introduced to the fuzzy logic technique used.

TABLE OF CONTENTS

DECLARATION

APPROVAL ii

DEDICATION iii

ACKNOWLEDGEMENT iv

TABLE OF CONTENTS

LIST OF TABLES viii

LIST OF FIGURES ix

ABSTRACT x

CHAPTER ONE: introduction 1

1 .0 Background to the study

I I Problem Statement 8

I .2 Aim and Objectives 9

1 .3 Research Questions 10

1 .4 Scope of Study 11

1 .5 Significance of the Study Ii

I .6 Thesis Organization II

CHAPTER TWO: Literature Review 13

2.0 Performance Indicators 13

2. 1 Knowledge representation technique 1 4

2.2 Review ofRelated Works 17

2.3 The fuzzy logic techniques 21

2.3.1 The uniqueness offuzzy logic concept 22

2.3.2 The benefit ofusing fuzzy logic approach 22

2.4 Basic Definitions 23

2.5 The concept of set theory 25

2.6 Mamdani’s Fuzzy Inference Method 30

2.7 Performance Indicators and Evaluation Criteria 31

2.8 Linguistic Variables 32

2.9 Determination of Membership Value 33

2.10 The Use ofMatlab Software 33

CHAPTER ThREE: Methodology 34

3.0 The Soft Computing Method 34

3.1 The Proposed Approach 34

3.2 Data Collection Methods 34

3.3 Modelling ofKPIs 35

3.4 The flow chart for the proposed study 40

3.5 The Procedures of Using the Fuzzy Logic Techniques 41

CHAPTER FOUR: Results and Discussion 47

4.0 Rules generation 47

4.1 The Membership Functions Generated 47

4.2 Model Validation 52

CHAPTER FIVE: Conclusion and Recommendations 54

5.0 Conclusion 54

5.1 Recommendations 55

Contribution to Knowledge 55

References 57

Appendices 60

Appendix 1: Resources Required 60