Prioritizing Risks
No program can mitigate all risk and uncertainty. Some risks may be highly probable yet cause a relatively small delay to the finish date. Conversely, a risk may potentially delay the program a long time but be highly unlikely to ever occur. In addition, it is impossible to fully mitigate uncertainty because of its inherent ambiguity. Therefore, regardless of the method used to examine schedule activity duration uncertainty, it is important to identify the risks that contribute most to the program schedule risk. These risks can then be targeted for mitigation strategies.
Sensitivity measures reflecting the correlation of the activities or the risks with the final schedule duration can be produced by most schedule risk software. Figure 39 shows a standard schedule sensitivity index for the house construction project.
Figure 39: Sensitivity Indexes for the House Construction Schedule
In the example in figure 39, setting the steel columns and beams affects the schedule duration more than digging the foundation, securing the construction loan, or the length of the approval period for the certificate of occupancy. The duration sensitivity chart identifies activities and paths that tend to be associated with project risk.
Figure 40 is a risk tornado chart, showing the correlation between a risk driver and project duration. It shows that when a risk is assigned to several activities, its sensitivity measure reflects the entire correlation, not just the correlation of one activity to the project’s duration. According to this analysis, incomplete design is the biggest risk driver in the house construction schedule, followed by the availability of materials. Using this information, the owner and general contractor can work together to ensure that the design is complete before the project begins, as well as identify alternative sources for key materials.
Figure 40: Evaluation of Risk Sensitivity in the House Construction Schedule
Risk analysis should also identify the activities that most often ended up on the critical path during the simulation, so that near risk-critical path activities can be identified and closely monitored. Risk criticality represents the percentage of simulation iterations that an activity or milestone is on the critical path. Figure 41 shows risk criticality for selected activities in the house construction schedule.
Figure 41: Risk Criticality of Selected Activities in the House Construction Schedule
In figure 41, the activities most likely to be on the critical path may not be the most risky themselves. The activities may be critical because they are appearing on a path whose criticality is driven by some risk affecting other activities.
Sensitivity indexes and correlation measures are useful starting points for assessing the possible magnitude of realized risks, but they have limited use in prioritizing risks. Notice that figure 39 shows activities, not the root cause risks. Thus, while the chart is useful for indicating where risk is the greatest, it cannot be used to identify specific risks for mitigation. And while figure 40 gives the correlations between risk drivers and project duration, no measure of time or cost is associated with the risks.
If the risk drivers method is used, the risks can be prioritized by their effect on the risk of finishing on time and their share of required contingency. If one risk at a time is removed and the Monte Carlo simulation is rerun, the contribution of each risk to the required contingency can be calculated at any percentile. The general process is to
Run the SRA using all risks and uncertainties. Record the finish date at the desired percentile—for example, 80 percent.
Remove a risk and run the SRA again. Compare the 80th percentile date with the date of the full model. The difference in the two dates is the expected contribution (or days saved) of the removed risk.
Replace the risk, remove the next risk, rerun to the SRA, and again compare the 80th percentile date to the date from the full SRA simulation and calculate the difference in days. Continue removing risks and uncertainties one by one; the risk with the highest contribution in days is the most important risk.
To identify the next most important risk, the most important risk identified above is removed from the model and the process is repeated. As the risks are removed and the next most important risks are identified, a prioritized list of risks and uncertainty can be created, similar to table 5. Table 5 shows the top 5 risks and their contribution to the 80th percentile date of April 23, 2026.
Table 5: Top Prioritized Risks in the House Construction Schedule
| Priority | Risk | 80th percentile date | Calendar days saved |
|---|---|---|---|
|
|
All | April 23, 2026 |
|
| 1 | Design is incomplete | April 1 | 22 |
| 2 | Machines are not available | March 23 | 9 |
| 3 | Material is not available | March 17 | 6 |
| 4 | Efficacy of the general contractor | March 10 | 7 |
| 5 | Materials are late or defective | March 4 | 6 |
Source: GAO | GAO-16-89G
The results in table 5 are the same as the risks identified in figure 40, although not in the same order. The order is likely to be different because as risks are removed from the simulation, activity durations and critical sequences change, depending on the nature of the schedule network and how risks and uncertainty were assigned to activities.