Assessing cost uncertainty: Lessons from environmental restoration projects

The uncertain physical and regulatory conditions prevalent in the field of environmental restoration often lead to considerable project cost growth. To date, the ability of project personnel to predict the magnitude of such growth has been limited. A formal approach to identifying, classifying, and incorporating uncertainty into standard cost estimating procedures is needed to quantify the risk of large-scale cost growth prior to project commencement. Because the elicitation of expert knowledge is critical to the risk analysis process, this paper evaluates the effectiveness of several approaches to eliciting and encoding probabilistic judgments. In essence we ask, "How should one frame questions concerning uncertain judgments in order to get the best answers?" In addition, this paper examines using both influence diagramming and Monte Carlo simulation to model the uncertainties associated with environmental restoration projects.