USING EXPERT JUDGEMENT TECHNIQUES TO ESTIMATE IT PROJECTS COMPLETION TIME

Abstract

Accurate estimation is crucial for the success of software development projects, but overestimation or underestimation of tasks remains a common issue. This article aims to explore the use of expert judgement techniques to improve IT project completion estimations. Expert judgement involves seeking input from experienced professionals to estimate the work required for a task. However, expert judgement is subject to biases and cognitive biases that can affect the accuracy of estimates. This research examines the causes of over and underestimation in software development projects and proposes practical solutions to reduce the risks associated with expert judgement estimation. The study utilizes a case study approach and compares single-point estimation with the Program Evaluation and Review Technique (PERT) to improve accuracy. The results demonstrate that PERT estimation provides a range of estimates with an expected case, and the average estimation is closer to the best case than the original estimates. Additionally, confidence intervals are calculated to provide a range of estimates at different confidence levels. The findings of the study demonstrate that PERT estimation, with its range of estimates and expected case calculation, improves estimation accuracy compared to single-point estimation. Additionally, the research calculates the Mean Relative Error (MRE) to measure the accuracy of estimates, indicating the potential for ongoing improvement in estimation accuracy. This article contributes to the field of IT project management by highlighting the importance of expert judgement techniques, identifying the challenges associated with over and underestimation, and providing practical tools and methodologies to enhance estimation accuracy. By minimizing subjectiveness and continuously improving estimates, IT projects can be delivered on time and within budget, leading to improved project outcomes and stakeholder satisfaction.