“Reference Class Forecasting” – databased risk estimation for complex projects

Abstract This publication presents the data‐based technique of Reference Class Forecasting for risk management in complex major projects. Complex projects consist of plannable and dynamic parts, which lead to changes, surprises and unknown events. Not all opportunities and risks can be estimated correctly. Reference Class Forecasting provides a risk uplift which, due to its statistical nature, also adequately considers feedback loop effects, human expertise and problem‐solving skills as well as the unknown. Reference classes for new projects are formed using a valid project evaluation and statistics on the risk surcharges required for completed projects. Assignment to a class allows a recommendation to the risk uplift.