Content area
Full Text
Risk analysis techniques have been used for years as one tool for estimating contingency. The definition of contingency advocated by AACE has recently been changed to reflect the use of statistical analysis to estimate the amount added to an estimate to allow for changes that experience shows will likely be required. Risk analysis techniques, like Monte Carlo simulation, are now accepted as the single tool or procedure for estimating contingency. Their usage increased as projects became more complex and computer technology became less expensive, allowing for more sophisticated methods of estimating this often misunderstood amount. Today there are several commercially available software products applying Monte Carlo risk analysis techniques to depict economic models including construction estimates.
Risk analysis can be used by owner companies to accomplish multiple objectives. First, management (or client) can establish guidelines for contingency. These guidelines should be used during the project's execution to draw-down the contingency and not just to develop the initial estimate. Second, management can establish overrun/underrun dollar limits where a project's authorization must be reviewed. Third, risk analysis advises management on the appropriate steps required to reduce the risk and corresponding overrun dollar limit if the project is marginally economical. Finally, management can establish performance measures for project management groups in companies where there are multiple projects based upon the number of projects being completed outside the approval range.
Using risk analysis to manage contingency offers several advantages. The steps necessary to develop the models improves communications within the project team and with the sponsoring organization or client. The ownership of contingency funds is not an issue because all participants are involved in the process and the amount of contingency in the forecast is a calculation, not a negotiation. Many techniques of managing contingency do not involve the project team, or they blindly follow rules established by the client when such techniques may not be appropriate.
ESTABLISHING CONTINGENCY
Several textbooks are available describing the technical aspects of Monte Carlo simulation and its use for project risk analysis. Simply stated, the simulation model has used a random number generator to build the project 1,000 or more times in the computer, storing the results after each iteration through the model. The mathematical model is built as a mirror of...