Using data you (probably) already have to predict program risks – Kwame
The traditional way to manage risks on projects calls for the project team to (1) identify risks, (2) log them, (3) analyze them to determine their likelihood and impact to the project, and (4) formulate the appropriate responses. In this presentation, I will attempt to show a better approach to addressing steps 3 and 4. I will show how we can apply the binomial distribution function to data collected and correctly categorized in a risk & issue database to predict which risks are likely to occur, and subsequently, how to setup appropriate contingency using normal distribution functions (or where sample sizes are too small, unified scheduling method or estimation by analogy). I will also show some important applications of these concepts including using calculated project risk scores to determine project “riskiness” and how to apply appropriate management based on this.
PMI Talent Triangle: Technical Project Management (Ways of Working)