This study aims to develop and validate an integrated optimization framework that simultaneously balances time-cost-quality trade-offs (TCQT) and resource leveling for construction projects, addressing a long-standing gap in project-scheduling research and practice.
To generate Pareto-optimal schedules, a non-dominated sorting genetic algorithm is implemented within the Excel-based SolveXL add-in, enabling multi-objective optimization directly within a practitioner-friendly platform. The resulting trade-off solutions are then ranked using the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS), guided by decision weights derived via the ordinal priority approach. This integrated framework combines evolutionary optimization with multi-criteria decision analysis to support transparent, stakeholder-informed schedule selection.
The algorithm produced non-dominated schedules. The best-compromise solution cut project duration from 81 days (most-likely baseline) to 47 days (42%), reduced total cost by 12% and lifted overall quality by 3.13%, while smoothing resource usage and halving the aggregate resource-moment index.
The validation relies on a single, deterministic case study. Broader generalization requires testing on larger, stochastic, and building information modeling (BIM)-derived datasets and inclusion of environmental and safety performance indicators.
By operating entirely within a familiar spreadsheet environment, the proposed model empowers planners to rapidly explore scenarios such as acceleration, crashing and crew reallocation. This capability not only supports data-driven decision-making in the field but also lays a flexible foundation for future research in artificial-intelligence-assisted project scheduling, BIM integration and sustainability modeling. Furthermore, the user-friendly framework enhances communication with stakeholders and promotes more transparent, accountable and optimized scheduling decisions, effectively bridging the gap between advanced research and practical construction management.
Smoothing resource demand reduces overtime peaks, supports safer working conditions, promotes fair labor allocation and curbs unnecessary equipment idling, contributing to lower emissions and better workforce well-being.
Unlike earlier studies that optimize either TCQT or resource leveling individually, this research unifies both objectives in a single GA-driven spreadsheet tool and is the first to embed OPA-weighted TOPSIS directly within the optimization loop. The result is an end-to-end decision-support framework that moves users seamlessly from the Pareto front to an actionable schedule inside a familiar Excel environment.
