A methodology for selecting portfolios of projects with interactions and under uncertainty

Abstract

Effective project evaluation and selection strategies can directly impact organizational productivity and profitability. Numerous analytical techniques ranging from simple weighted scoring to complex mathematical programming approaches have been proposed to solve these problems. However, traditional project selection methods too often fail to consider both the uncertainties in projects and the interaction among projects. Some prior studies have considered the interaction among projects in deterministic environments. Others have dealt with stochastic environments but have not considered project interdependencies. This study aspires to fill this gap in the project portfolio selection literature. Information system/information technology (IS/IT) projects are used in this study because they are frequently subject to uncertainties due to estimation difficulties and bounded by interactions due to technological interdependencies. We use Data Envelopment Analysis (DEA) to select the best portfolio of IS/IT projects while taking both project uncertainties (modeled as fuzzy variables) and project interactions into consideration simultaneously. We also present a numerical example to demonstrate the applicability of the proposed framework and exhibit the efficacy of the procedures.