Project portfolio selection with relationship considerations: A random walk with restart based on influencing factors

Selecting an optimal project portfolio (PP) is a critical yet challenging task for organizations in an increasingly diverse and complex environment. The dynamic and holistic impacts of diversity and complexity on PP strategic goals help to reflect the complex realities of project portfolio selection (PPS). However, traditional research on PPS fails to adequately consider the diversity of influencing factors and the complexity of relationships among projects, among influencing factors, as well as between projects and influencing factors. To address the gaps, a novel PPS methodology was proposed by integrating a two-layer network model with a random walk with restart algorithm. First, a two-layer network containing a project layer and an influencing factor layer was constructed to portray the impacts of factors on the implementation of PP. Second, a random walk with restart was employed to capture the changes of project contribution to the goal under the complex synergy relationships among projects, among influencing factors, and between influencing factors and projects. Finally, a detailed case study verified the viability of the proposed method. The main findings of this study were in two aspects: (1) it was necessary to consider complex relationships for PPS under uncertain and dynamic environments. (2) the proposed method enabled better alignment of selected PP with strategic goals by incorporating these complex interdependencies. This study theoretically facilitates the PPS with the consideration of integrated three complex relationships that exist in PP. Practically, this study offers actionable tools directly applied to improve strategic decision-making in complex project environments.