Risk-Informed Ranking of Engineering Projects
9th International Conference on CANDU® Maintenance - 2011 December 04-06

Presented at:
9th International Conference on CANDU® Maintenance
2011 December 04-06
Toronto, Canada
Session Title:
Industry Performance Special Session C1

Mikko Jyrkama (University of Waterloo)
Mahesh Pandey (University of Waterloo)


Refurbishment planning requires prudent investment decisions with respect to the various systems and components at the station.  These decisions are influenced by many factors, including engineering, safety, regulatory, economic, and political constraints.  From an engineering perspective, the concept of cost-benefit analysis is a common way to allocate capital among various projects.  Naturally, the “best” or optimal project should have the lowest cost and the highest benefit.  In the context of risk-informed decision making (RIDM), a process that has been widely embraced by the global nuclear community, the costs and benefits must further be “weighted” by probabilities to estimate the underlying risk associated with the various planning alternatives.

The main purpose of this study is to illustrate how risk and reliability information can be integrated into the refurbishment planning process to facilitate more objective and transparent investment decisions.  The methodology is based on the concept of generation risk assessment (GRA) which provides a systematic approach for balancing investment costs with the reduction in overall financial risk.  In addition to reliability predictions, the model provides estimates for the level of risk reduction associated with each system/project and also the break-even point for investment.  This information is vital for project ranking, and helps to address the key question of whether capital investment should be made in the most risk critical systems, or in systems that reduce the overall risk the most. 

The application of the proposed methodology requires only basic information regarding the current reliability of each engineering system, which should be readily available from plant records and routine condition assessments.  Because the methodology can be readily implemented in a Microsoft Excel spreadsheet, all plausible (e.g., bounding) planning scenarios, with or without investment, can also be generated quickly and easily, while at the same time using sound probabilistic principles and models.  Sensitivity analysis can also be performed instantly, by changing key model parameters, such as escalation factors, failure costs, repair times, etc.  The uncertainty with each project can further be quantified through the concepts of value-at-risk (VaR) and expected shortfall (ES) given uncertainties in the model input parameters.  

The benefits of the developed methodology are demonstrated by a simple example application involving a small number of competing projects.

Individual Conference-Paper Copies (Electronic Where Available):

  • For CNS members, the first 5 copies per calendar year are free, and additional copies are $10 each.
  • For non-members, the price is $25 for the first Conference-paper copy in a request, and $10 each for additional copies of papers in the same conference and in the same request.
  • Contact the CNS office to order reprints.