Estimating Cost of Software Development with Traditional and Hesitant Fuzzy Pairwise Comparison Methods: An Application

Estimating the cost of software is a fundamental and difficult process in information technology (IT) industry. Underestimation causes assigning fewer resources than the project needs and setting an unrealistic schedule. On the other hand, overestimation results in waste of resources and low customer satisfaction. It is clear that, software projects can be prioritized efficiently and resources can be used effectively by using efficient estimation methods. This paper presents a Traditional Pairwise Comparison (TPC) which involves a number of factors selected with the help of expert judgments working in a Turkish IT company. There may be uncertainties while evaluating the relative importance of two factors with TPC. Therefore, Hesitant Fuzzy Pairwise Comparison (HFPC) using Hesitant Multiplicative Geometric (HMG) operator is also presented to estimate cost of software projects having uncertainty in judgment. Both methods are applied to a software cost estimation problem for a Turkish IT company. It is seen that both methods provide efficient estimations in comparison with the real effort.