PM 587 Week 2 Qualitative Risk Analysis Example

Thirty Risks, Ten That Matter Now: A Qualitative Risk Analysis for a Composite Panel Launch

[Student Name]

University of Phoenix

PM/587: Project Risk Management and Quality Assurance

Week 2 Assignment

[Instructor Name]

[Date]

Plainsview Aerostructures, its risks, scales and scores are composites written for a model paper.

Week 1 identified thirty risks for Plainsview Aerostructures' launch of 22 carbon-fiber panels for a new business jet, a composite project in Wichita with first article inspection due in month 13 and first delivery in month 14. A list of thirty risks is not yet a plan. The team needs to know which risks deserve attention now, which can be watched and which need deeper analysis. This paper explains how the team analyzed and ranked them.

Scales Built From Objectives

The current standard notes that qualitative analysis assesses the probability and impact of individual risks using defined scales, so that risks can be prioritized for further action (Project Management Institute [PMI], 2021). Undefined scales invite each scorer to imagine different meanings, so the team wrote scales in the project's units.

Probability: 1, under 10 percent; 2, 10 to 30 percent; 3, 31 to 50 percent; 4, 51 to 70 percent; 5, over 70 percent.

Impact, using the worst of three dimensions: 1, under one week of delay to first article inspection, under $50,000 or under one point of yield; 2, one to two weeks, $50,000 to $150,000 or one to two points; 3, two to four weeks, $150,000 to $400,000 or two to four points; 4, four to eight weeks, $400,000 to $1 million or four to eight points; 5, more than eight weeks, more than $1 million or more than eight points.

Opportunities used the same scales in the positive direction.

The Scoring Workshop

Twelve people scored the risks in a half-day workshop. Tversky and Kahneman (1974) showed that people estimating uncertain values lean on mental shortcuts, sticking close to the first number they hear and rating events as likely when similar cases are easy to recall, and that these shortcuts produce predictable errors. To reduce anchoring, each participant first scored every risk silently on a form. Scores were then displayed together, and the group discussed only those risks where individual scores differed by two or more levels, about a third of the list. The facilitator asked the people with outlying scores to explain their reasoning first, so that minority information was heard before the group converged.

The Ranked List

Risk scores were calculated as probability multiplied by impact. The top ten:

Late customer drawings for six parts: probability 4, impact 4, score 16.

Autoclave slots unavailable during qualification: probability 4, impact 4, score 16.

Prepreg supply cut from the single qualified supplier: probability 3, impact 5, score 15.

Invar tool delivery slips beyond 22 weeks: probability 3, impact 4, score 12.

Prepreg out-time exceeded on long layups: probability 3, impact 4, score 12.

Coordinate measuring machine bottleneck during inspection: probability 4, impact 3, score 12.

First-pass yield below 92 percent by month 20: probability 3, impact 4, score 12.

Technician hiring shortfall: probability 3, impact 3, score 9.

Customer source inspector rejects report format: probability 2, impact 4, score 8.

Loss of the ply-cutter programmer: probability 2, impact 4, score 8.

The top opportunity, early completion of the military lot freeing autoclave time, scored probability 3 and impact 3.

A score of 16 does not mean twice as dangerous as 8; it means look here first.

Proximity and Confidence

Two more ratings sharpen the list. Proximity rates how soon a risk could occur: near (within three months), medium (three to nine months) or far. Late drawings and tool delivery are near; yield and hiring are medium. Confidence rates the information behind each score: high, medium or low. The prepreg risk has medium confidence, since the supplier's allocation history is known but its future capacity is not; the yield risk has low confidence, since the new shapes have no history. Low confidence on a high-scoring risk is a reason for more analysis, not less concern.

Known Weaknesses of Risk Matrices

Thomas et al. (2014) examined risk matrices used in the oil and gas industry and showed that their rankings can change with seemingly minor design choices, such as where scale boundaries fall, and can lead to decisions that differ from those a quantitative analysis would support. Plainsview's team saw this in practice: the prepreg risk scored 15 and the drawings risk 16, yet a prepreg cut could stop all production while late drawings affect six parts. The team therefore treats scores as a sorting tool and reviews the top risks' consequences directly rather than relying on the numbers alone.

Risks That Move Together

Some risks are linked, and a matrix scores them one at a time. Late customer drawings and invar tool delivery interact: if drawings arrive late, tool design starts late, which pushes the tool order and makes the 22-week lead time more dangerous. Autoclave conflict and prepreg out-time also interact, since waiting for an autoclave slot is one of the main reasons prepreg sits too long on the layup table. The team drew a simple influence diagram linking these pairs and noted that responding to the upstream risk, drawings and autoclave slots, would lower the downstream ones. This is another reason the linked risks go to quantitative analysis, where their combined effect on the schedule can be modeled.

Scoring Opportunities

Opportunities were scored on the same scales, with impact read as weeks saved, dollars saved or yield points gained. Early completion of the military lot scored 3 and 3. Qualifying the newer out-of-autoclave material scored lower on probability, 2, because the customer would have to approve a material change, but higher on impact, 4, since it would remove much of the autoclave constraint for the smaller panels. Automated ply cutting scored 4 on probability, since the machine is already installed, and 2 on impact.

What Happens Next

Risks were placed into three groups. Act now: late drawings, autoclave slots and tool delivery, all near-term with high scores, receive response plans immediately. Analyze quantitatively: prepreg supply, out-time, inspection bottleneck, yield and the autoclave opportunity go into a Monte Carlo schedule and cost model in Week 3, because their effects interact and their confidence ratings are not high. Watch: the remaining twenty risks stay on the register with triggers and monthly review.

Who Owns the Top Risks

Each of the top ten keeps the owner named in Week 1, and the ranking adds a deadline: owners of act-now risks must bring a response plan to the next weekly launch meeting, while owners of risks bound for quantitative analysis must supply three-point estimates of their schedule and cost effects within two weeks.

Reviewing the Scores

Scores are not permanent. The team will rescore the register monthly, when customer drawings are released and after the first qualification panels are cured. A risk whose proximity moves from medium to near is automatically discussed at the next review.

Conclusion

Qualitative analysis turned thirty identified risks into a ranked, time-aware list. Scales written in the project's own units, silent individual scoring before discussion, proximity and confidence ratings and a frank view of the matrix's weaknesses gave the launch team a defensible basis for action. Three risks receive responses now, five go to quantitative analysis and the rest are watched, which is how a register becomes a management tool.

References

Project Management Institute. (2021). A guide to the project management body of knowledge (PMBOK guide) (7th ed.). Project Management Institute.

Thomas, P., Bratvold, R. B., & Bickel, J. E. (2014). The risk of using risk matrices. SPE Economics & Management, 6(2), 56-66. https://doi.org/10.2118/166269-PA

Tversky, A., & Kahneman, D. (1974). Judgment under uncertainty: Heuristics and biases. Science, 185(4157), 1124-1131. https://doi.org/10.1126/science.185.4157.1124