Project Portfolio Risk Response Selection Using Bayesian Belief Networks

Abstract

Risk identification, impact assessment, and response planning constitute three building blocks of project risk management. Correspondingly, three types of interactions could be envisioned between risks, between impacts of several risks on a portfolio component, and between several responses. While the interdependency of risks is a well-recognized issue, the other two types of interactions remain unacknowledged in the risk response planning literature. This research suggests a Bayesian belief network for modeling portfolio risks, their impacts, and responses. There are three kinds of nodes in this network: nodes representing portfolio risks, nodes corresponding to risk impacts on each objective of each portfolio component, and nodes showing response actions. The problem is to decide which responses are to be selected. For this purpose, an optimization model is proposed that minimizes the sum of both residual risk effects on portfolio component objectives and response implementation costs. Subsequently, a genetic algorithm is introduced to solve the model. A simple portfolio instance is also provided to illustrate the proposed model.