Portfolio management in the construction sector represents a key issue with an impact on strategic development and competitiveness, involving the consideration of a large number of items, increasing its complexity. Permeated not only by financial issues, but also by ecological ones, the selection and management of portfolios should consider long-term organizational strategic objectives, so traditional measures based on value alone become inadequate. Thus, integrating partial information and the benefit-to-cost heuristic, this paper proposes a Multi-Criteria Decision Making/Aiding (MCDM/A) model to support Decision-Makers (DMs) in such problems, making companies less susceptible to risks despite the higher number of items.
In this study, a framework with two phases is applied. In the first, the problem is structured, defining the main objectives for the problem and the related criteria. In the second phase, the MCDM/A model is built through the FITradeoff method to obtain a portfolio of items in accordance with preference information and the budget. A case study illustrates the application of both the framework and the proposed model.
Based on objectives that encompass not only financial aspects but also ecological and strategic ones, six evaluation criteria were defined. Using these criteria, 147 projects from a construction company in Brazil were assessed. A portfolio with 68 alternatives was then developed, utilizing a simplified heuristic structure that ensured process agility and required only partial information about the DM’s preferences, which contributed to the simplification of the process. At the same time, this approach maintained a solid axiomatic foundation, providing robustness to the developed model. The results achieved were satisfactory for the company and were validated by the DM.
This work is the first to demonstrate the feasibility and success of the benefit-to-cost heuristic approach with partial information applied to a portfolio problem with a large number of projects. Additionally, the advantage of using incomplete information is highlighted, as this allows DMs to obtain partial results from the FITradeoff method. In addition, this study contributes by offering a model that provides suggestions for criteria that are consistent with the context in which the construction sector operates, extending beyond the purely economic perspective.
