Multi-factor association dependence modelling for project risk analysis

Research output: Contribution to journalArticlepeer-review

Abstract

This article presents an analytic (non-simulation) dependence model for quantitative project risk analysis. The multi-factor association model (MFAM) accounts for multiple association factors in typical project settings and provides a closed-form solution to a complete and mathematically consistent correlation matrix. Given standardized, ubiquitous project plans (e.g., work breakdown structure, resource allocation, or risk register), the MFAM establishes a hierarchical tree of association factors, which is subsequently encoded into an analytic model for quantitative risk analysis. In this article, we present the MFAM programmed in Microsoft Excel and demonstrate the computational efficiency of the MFAM using two alternative schedules with distinct resource utilizations. With the enhanced robustness in dealing with multiple risk factors in a project and the computational efficiency from the non-simulation second-moment approach, the MFAM offers additional flexibility for and scalability to large-scale project risk analysis problems. The modelling procedures and solutions presented in this article highlight three potentials of the MFAM as a robust quantitative risk analysis tool. • The MFAM can be fully programmed in Microsoft Excel using only basic cell functions. • Under mild assumptions, the MFAM provides reliable risk estimates comparable to Monte Carlo simulation. • The MFAM is scalable to high-dimensional risk problems (i.e., with multi-thousands or more) with an affordable computational burden.

Original languageEnglish (US)
Article number101443
JournalMethodsX
Volume8
DOIs
StatePublished - Jan 2021

All Science Journal Classification (ASJC) codes

  • Clinical Biochemistry
  • Medical Laboratory Technology

Cite this