1. Assume that a project is to design a high-speed cellular modem for release to the customer, IBM, on or before July 30. Based on experience with other product design projects,knowledge of the current supplier base, and using the schedule based approach the project team has determined that the following twelve possible failure modes should be part of the projectrisk analysis.
2. F1. The supplier will fail to deliver the needed components and assemblies on time to meet in-house production dates. Based on experience with this supplier, the probability is quite high that delivery will be late and the probability is low that the team will find out about it in advance. Further, if it does happen, the impact on the project is likely to be very high.
3. F2. The supplier will fail to meet product specifications. Experience shows that the risk is medium to high that this will occur. If the supplier fails to meet specifications, however, there is a high likelihood that it will be caught early. If the discrepancy is not caught, the impact on the project will be very high.
4. F3. The supplier will miss cost targets. This is a medium to high probability with a medium to a high likelihood that the overrun will be detected in advance. The impact on margin will be high if it does occur, however.
5. F4. The product fails reliability testing. This is a low to medium probability given the supplier's quality systems and SPC techniques. Nevertheless, if it does occur, the impact will be moderate due to the robustness that was built in. The bad news is that the probability of detection is low.
6. F5. The product fails accelerated life testing. This, too, is a low probability with a moderate impact should it occur. Again, the level of robustness that was built in should handle moderate swings in product characteristics. As before, there is a low probability of detecting it in advance. Maybe not knowing is good?
7. F6. The product could prove to be unproducible during prototyping. The likelihood of detection of this failure is medium to high based upon the design of experiments (DOE) and repeatability and reliability (R&R) analyses, but still of concern. The probability of occurrence is low and the impact minor because there would be time to react prior to production, and there is a parallel design in the works just in case.
8. F7. Internal product cost targets could be missed. Experience shows the probability of this occurrence is high. There is a poor record of accurately predicting internal costs. Detection is medium to high, but probably on the lower range of the scale. The impact could be moderate, as there is a significant margin built in just in case.
9. F8. Production capacity may not be sufficient to meet the actual demand for the product. Based on the ability to go to a third shift, the impact of this would likely be low, and there would be a high likelihood of detecting it with sufficient warning to make the necessary arrangements. And, there is enough experience in this area that the probability is only low to medium that it would occur.
10. F9. The product might not be completed in time for the scheduled prototype run. Here, the probability is very high given past problems with production and its associated tooling. The problem is that the probability of detecting the failure is low. The impact on the project will be moderate because additional resources will be thrown at the problem to get back to the scheduled production dates.
11. F10. The product will not be ready for production as scheduled. Now, this is a big one, with a very high likelihood that it could occur. If it does, the impact on the project will be severe. There is a high likelihood that it will be detected in advance, but it would still make things unpleasant.
12. F11. The sales and marketing literature will not be ready on time to support shipments. This happens occasionally and the impact is typically low because there is the ability to respond quickly with in-house publications staff. There is a high likelihood that a miss in this area will be detected.
13. F12. The price books will not be ready for distribution to the retail outlets prior to the initiation of shipments as requested. This, too, happens occasionally and even though the impact on the project is only moderate, the retailers will not be happy about it. However, there is a good chance that the failure would be detected with sufficient warning.
14. Use the following table to quantifying the risk

Probability of Occurrence (P.) Quantitive Value Very high (occurrence predictable) l in 2 High (repeated failures) 1 in 40 Medium to high 1 in 400 Medium (occasional failures) 1 in 1,000 Low to medium 1 in 4,000 Low 1 in 20,000 Remote (occurrence unlikely) 1 in 1,000,000 Probability of Detection (PD) Quantitive Value Remote (detection impossible) Less than 1% Very low (detection unlikely) 1% to 10% Low 10% to 30% Medium (detection possible) 30% to 50% Medium to high 50% to 70% High (good chance of detection) 70% to 90% Very high (detection predictable) 90% to 100% Magnitude of Failure Effect (MF) Quantitive Value Very high (failure of the project in the customer's eyes a given) 9,10 High (high degree of customer dissatisfaction probable) 7,8 Moderate (customer wil experience a moderate level of dissatisfaction) 4,5,6 Low (slight effecto only) 2,3 Minor (no measureable effect) 1