Enhancing project portfolio selection for construction holding firms: a multi-objective optimization framework with risk analysis
Milad Ghanbari,Asaad Azeez Jaber Olaikhan,Martin Skitmore
DOI: https://doi.org/10.1108/ecam-05-2023-0532
2024-06-08
Engineering Construction & Architectural Management
Abstract:Purpose This study aims to develop a framework for the optimal selection of construction project portfolios for a construction holding company. The objective is to minimize risks, align the portfolio with the organization's strategic objectives and maximize portfolio returns and net present value (NPV). Design/methodology/approach The study develops a multi-objective genetic algorithm approach to optimize the portfolio selection process. The construction company's portfolio is categorized into four main classes: water projects, building projects, road projects and healthcare projects. A mathematical model is developed, and a genetic algorithm is implemented using MATLAB software. Data from a construction holding company in Iraq, including budget and candidate projects, are used as a case study. Findings The case study results show that out of the 34 candidate projects, 13 have been recommended for execution. These selected projects span different portfolio classes, such as water, building, road and healthcare projects. The total budget required for executing the selected projects is $64.55m, within the organization's budget limit. The convergence diagram of the genetic algorithm indicates that the best solutions were achieved around generation 20 and further improved from generation 60 onwards. Practical implications The study introduces a specialized framework for project portfolio management in the construction industry, focusing on risk management and strategic alignment. It uses a multi-objective genetic algorithm and risk analysis to minimize risks, increase returns and improve portfolio performance. The case study validates its practical applicability. Originality/value This study contributes to project portfolio management by developing a framework specifically tailored for construction holding companies. Integrating a multi-objective genetic algorithm allows for a comprehensive optimization process, taking into account various objectives, including portfolio returns, NPV, risk reduction and strategic alignment. The case study application provides practical insights and validates the effectiveness of the proposed framework in a real-world setting.
engineering, civil, industrial,management