Scenario-based optimization robust model project portfolio selection under risk considerations
Ramedani, Amir Ali,Didehkhani, Hosein,Mehrabian, Ahmad
DOI: https://doi.org/10.1007/s00521-022-07434-8
2022-10-07
Neural Computing and Applications
Abstract:In project portfolio selection (PPS) management, one of the main goals is the optimal management of projects with the least risk and the highest commercial value under risk considerations. Hence, this study considers the weight of each decision criterion, their impacts, and also the uncertainty in decision making. By taking into account all those assumptions, this paper seeks to conduct a PPS with aiming of maximizing the average value as the performance of each project, the rate of development of each project and minimizing the risk of interruption in the implementation of selected projects. The strategic goal of this study is to select robust project portfolios in the long run for less replacement. Accordingly, for attaining all goals, a combined method developed in three stages of PPS; first the weight of criteria from the F-AHP method is determined, next the F-TOPSIS method is used to calculate the relative scores for the projects, and finally a scenario-based robust multi-objective mathematical programming model is considered. This paper has been encountered with two challenges and complexity which is solved by the hybrid method based on the Multi-Choice Goal Programming with Utility Function (MCGP-UF) and the particle swarm optimization (PSO) algorithm (hybrid PSO-MCGP-UF). The results show an improvement in the solution time and the quality of the responses of the proposed method, which helps decision-makers at all stages of the PPS to achieve robustness portfolios in less time.
computer science, artificial intelligence