A comparison of risk analysis approaches

Abstract

Risk management is undoubtedly a precious component of the overall project management field. It assists in identifying, assessing and mitigating the uncertainties that are present in a project. These uncertainties affect important measures of a project, of which cost and schedule are the most prominent, and consequently put project success under pressure. Risk analysis, especially in its quantitative form, is a core process of risk management that provides valuable insights regarding these uncertainties. Numerous different procedures for quantitative risk analysis exist, but generally, they can be classified in two types based on the way they treat the cost and schedule components of a project. In the Separated Approach two risk analyses are performed, one for schedule and one for cost, which are independent of each other. In the Integrated Approach one risk analysis is performed, where schedule and cost components are simultaneously analyzed, and are no longer independent of each other. The advantages and disadvantages for each approach are well described in literature. However, there is little empirical evidence of the performance of different approaches when they are applied on real projects. This research tries to fill this gap by comparing the outcomes of different approaches when they are implemented on a real project, against the actual outcomes of the project to get an indication of their relative performance. Particularly, the Proposed Separate Approach, the Proposed Integrated Approach and the Company’s Separated Approach are applied on a completed project and their outcomes regarding cost and schedule contingency are compared to each other and to the actual project outcomes. Based on this comparison, a conclusion is drawn, and an answer is given on which approach is more accurate. Regarding cost contingency estimate, the Proposed Integrated Approach was the more accurate, while for schedule contingency estimate, the Company’s Separated Approach was the most accurate. A similar research should be conducted on a large number of projects to be able to generalize this conclusion though.