GQM's AI-enhanced framework for proactive forecasting and financial assessment of software project delays

Software Project Delay, Cost Estimation, GoalQuestion-Metric, Artificial Intelligence, Machine Learning, Project Management, Integrated DelayCost Equation.

Ahmad Abdullah Alghamdi1*
General Directorate of Health Services, Ministry
of Defense, Saudi Arabia.
ORCID: 0009-0009-3390-4149
E-mail: ahmad.a.alghamdi@modhs.med.sa

Reem Alnanih2
Department of Computer Science, Faculty of
Computing and Information Technology, King
Abdulaziz University, Jeddah, Saudi Arabia,
Software Engineering and Distributed System
Research Group, King Abdulaziz University,
Jeddah 21589, Saudi Arabia.
ORCID: 0000-0002-2428-0356
E-mail: ralnanih@kau.edu.sa

Received:29-05-2026
Revised:27-06-2026
Accepted: 04-07-2026

European Journal of Prosthodontics and Restorative Dentistry (2026) 34(5s) 01–16

GQM's AI-enhanced framework for proactive forecasting and financial assessment of software project delays

Software project delays remain a persistent source of cost escalation, delivery uncertainty, and managerial risk, particularly in dynamic environments where requirements, resources, dependencies, and stakeholder decisions evolve throughout the project lifecycle. Conventional estimation approaches often rely on static assumptions and therefore provide limited support for early delay detection or financial impact assessment. This study develops and validates an artificial intelligence-enhanced Goal-Question-Metric framework, referred to as AI-GQM, for proactive forecasting and financial quantification of software project delays. The framework integrates a dynamic GQM-D measurement structure, machine learning-based prediction, and an Integrated Delay-Cost Equation to convert delay indicators into a financially interpretable risk estimate. A positivist quantitative design was adopted using three empirical sources: a global dataset of 10,000 completed software projects for model training, a separate Saudi sample of 52 completed projects for unseen contextual testing, and survey responses from 111 practitioners for indicator validation and weighting. Random Forest, Long Short-Term Memory, and XGBoost models were evaluated within the proposed framework. The AI-GQM framework achieved 92.5% predictive accuracy and demonstrated strong explanatory power for cost overrun estimation (R² = 0.892). The Integrated Delay-Cost Equation further supported proactive financial risk interpretation by linking delay severity, inter-factor dependency, prediction outputs, and projectspecific risk adjustments. The findings indicate that AI-GQM can improve delay forecasting, cost-risk visibility, and decision support in software project management.

•••••••••••••••••••••••••••••••• ejprd.org - Published by Riset Publication Services LLC

EJPRD

Copyright ©2026 by Riset Publication Services LLC