A novel mathematical programming model for multi-mode project portfolio selection and scheduling with flexible resources and due dates under interval-valued fuzzy random uncertainty

Seyed Meysam Mousavi | S. Zolfaghari

---

URL :   http://research.shahed.ac.ir/WSR/WebPages/Report/PaperView.aspx?PaperID=158799
Date :  2021/07/09
Publish in :    Expert Systems With Applications

Link :  https://www.sciencedirect.com/science/article/abs/pii/S0957417421006400
Keywords :Project portfolio selection, Project portfolio scheduling, Flexible resources, Multi-mode resource-constraints, Fuzzy random variable, Chance-constraint programming, Credibility-based programming

Abstract :

---

Resource management plays a pivotal role in project implementation success. The profitability of project-based companies not only relies on the right selection of project portfolio but also depends on accurate scheduling and resource management. This issue is addressed by optimal project portfolio selection and scheduling using flexible resource availability and projects’ due date. The motivation of flexible resources is to maximize cash flow by resource management and adding more projects to a portfolio by reinvesting the gain to increase resource availability during planning horizon. Flexible due date leads to perform more projects besides increasing the robustness of multiple projects. This study proposes a new linear structure of mixed-integer programming (MIP) model of project portfolio selection and scheduling considering the resource management, cash flow, tardiness cost, and robustness of multiple projects. Moreover, to a better description of real-life project situations, a new solution approach is introduced based on triangular interval-valued fuzzy random variables to incorporate both fuzziness and randomness uncertainties into mathematical programming models. To solve the proposed intervalvalued fuzzy random model, a generalized credibility-based chance constraint measure is applied. The application and performance of proposed model are evaluated through different small to large-sized test problems and a real case study to tackle uncertainties and complexities explicitly.