Info

timating probability and consequence of occurrence. Here, ordinal scales (scales whose values are rank ordered) are commonly used to represent different aspects of the probability of occurrence (e.g., due to technology design, or manufacturing) and consequence of occurrence (e.g., cost, schedule, and technical). While such scales, tailored to your project, can be a useful methodology for estimating risk, great care must be taken in using them. A common abuse of such probability and consequence scales is performing mathematical operations on the results, which can easily lead to erroneous results because the true scale interval values are unknown (e.g., a five level scale labeled 0.2, 0.4, 0.6, 0.8, and 1.0 almost certainly does not have values of equal 0.2 increments between adjacent scale levels).

The following simple example illustrates the proper use of ordinal scales in project risk analysis, and provides some recommendations for properly representing the results.6 Please note, these scales should not be used on your project—they are only provided as an illustration.

6. This example is condensed from Edmund H. Conrow, Effective Risk Management: Some Keys To Success (Reston, VA: American Institute of Aeronautics and Astronautics, 2000), pp. 301-305, Copyright ©2000, Edmund H. Conrow. Used with permission of the author.

Example 17-2. A single "probability" of occurrence scale, related to technology maturity, is used, as shown in Table 17-5. (Note: Since ordinal probability scales almost never represent true probability, but only an indicator of probability, I indicate scores derived from such scales as "probability" values.) In reality, technical risk will typically encompass a number of additional risk categories in addition to technology maturity, such as design, and so on. However, the use of a single risk category simplifies subsequent computations and is sufficient for illustration purposes. For the technology maturity "probability" scale, assume that low = scale levels A and B, medium = scale level C and D, and high = scale level E. (Note: This information does not correspond to low, medium, and high risk, and is only an indicator of where breakpoints will occur when used in developing the risk mapping matrix later in this section. Letters are provided for scale levels instead of numbers to discourage you from attempting to perform invalid mathematical operations on the results.)

Three consequence of occurrence scales, for cost, schedule, and technical, are used and given in Table 17-6. For each of the three consequence of occurrence scales, assume that low = scale levels A and B, medium = scale levels C and D, and high = scale level E. (Note: This information does not correspond to low, medium, and high risk, and is only an indicator of where breakpoints will occur when used in developing the risk mapping matrix later in this section.)

Given the mapping information associated with the "probability" of occurrence and consequence of occurrence scales, a mapping matrix was developed and is given in Table 17-7. [Note: Setting risk boundaries is often not exact since three divisions were used for both the "probability" and consequence of occurrence scales versus the five possible levels (one per scale level); a mapping matrix with different "probability" of occurrence and/or consequence of occurrence relationships (e.g., low = scale levels A and B, medium = scale level C, and high = scale levels D and E for both "probability" and consequence of occurrence scores), or five resulting risk levels (low, low medium, medium, medium high, and high), or different risk boundaries could also have been used for this example.]

We'll now evaluate two different items associated with a commercial high-grade digital camera, using the above risk analysis methodology. Remember, these risk issues are hypothetical and only used to illustrate how to apply the risk analysis methodology.

In the first case, a high-performance commercial charge-coupled device (CCD) exists that is in preprototype development. The CCD will be included in a high-grade digital

TABLE 17-5. EXAMPLE OF ORDINAL TECHNOLOGY "PROBABILITY" MATURITY SCALE

Definition

Scale Level

Basic principles observed

Concept design analyzed for performance

Breadboard or brassboard validation in relevant 