Quantum Machine Learning for Multi-Criteria Decision-Making in Industrial Project Portfolio Management

Abstract

Industrial project portfolio management (PPM) increasingly relies on complex, multidimensional decision-making processes that must balance financial constraints, risk profiles, resource limitations, sustainability targets, and strategic priorities. Traditional analytical and machine learning methods struggle to scale efficiently as portfolio complexity increases, particularly when decisions involve nonlinear interactions, high-dimensional criteria, and combinatorial optimization. This study explores the use of quantum machine learning (QML) frameworks to enhance multi-criteria decision-making (MCDM) in industrial PPM. By leveraging quantum-enhanced feature spaces, variational quantum classifiers, and hybrid quantum–classical optimization methods, the approach enables improved evaluation of project alternatives, more robust prioritization under uncertainty, and more efficient exploration of exponentially large decision spaces. The research demonstrates how QML-driven MCDM architectures can support dynamic portfolio balancing, scenario-based investment planning, and real-time reallocation strategies in fast-evolving industrial environments. The findings highlight the potential for quantum machine learning to shape the next generation of intelligent portfolio decision-support systems.