Estimating Project Completion Times using Fuzzy Logic: A Concurrent Engineering Project Management Application


Jarrod Rogers and Rob Rucker

Computer Integrated Manufacturing Center and

Industrial and Management Systems Engineering Department

Arizona State University

Tempe Az. 85287- 5106

Dec. 4, 1995


ABSTRACT

Estimating project completion time is a painstaking, tedious task which must be done in many different types of jobs in many different companies. Managers often want up to date and accurate estimates of when the customer can be promised the job will be done. The customer never wants an underestimate, but the estimate must also be reasonably accurate. The combined work of a project management system and fuzzy logic can save everybody a lot of time while giving accurate real-time estimates of project completion. The project management system collects large amounts of data directly from the source, the estimators. The project management system cuts out the middle people and saves massive amounts of time which would had been spent on data collection. All of this data is wonderful but if you can't do something constructive with it, it is useless. The estimation process involves human reasoning, human perception, and human decision making, and this is where fuzzy logic has the most success. The fuzzy logic model will reflect common human characteristics in making estimates. The fuzzy model helps quantify human estimates, giving mathematical support for improvement of these estimates. Fuzzy logic filters the information in the human estimates that otherwise would have been tossed out altogether. With a project management system and fuzzy logic, the job of project completion time could be a quick painless task done entirely by one person.

*This research is funded by the National Science Foundation, # 9215514, and conducted in the A.S.U. product development laboratory


TABLE OF CONTENTS

LIST OF TABLES

LIST OF FIGURES

CHAPTER

1.0 Introduction
1.1 The CEPM Solution
1.2 Why Fuzzy Logic
2.0 The Fuzzy Model
2.1 Fuzzy Variable
2.1.1 Hours Variable
2.1.2 Percentage Variable
2.1.3 Accuracy Variable
2.2 Fuzzy Rules
3.0 How the Fuzzy Model Works
3.1 Fuzzifying the Estimate
3.2 Rule Firing
3.3 Rule Interpretation
3.4 De-fuzzifying the Answer
4.0 Applying the Fuzzy Model to Project Completion
4.1 Breaking Up the Project
4.2 Summing Project Parts
5.0 Summary
References


[Home]   |   [Résumé]   |   [Employment History]   |   [Thesis]

[Email]   |   [Page]