Solution selection matrix (Pugh matrix)
After root causes are confirmed in the Analyze phase, the Improve phase requires selecting the best solution from multiple candidates. A solution selection matrix (a weighted criteria matrix; the Pugh matrix, below, is a related variant that compares options to a baseline) makes this choice transparent, objective, and defensible by scoring each option against defined criteria with assigned weights.
When to use a solution selection matrix
- Multiple feasible solutions exist and the best choice is not obvious
- Criteria pull in different directions (e.g., a solution with high effectiveness has high cost)
- The team needs to document and defend the rationale for their choice at the Improve tollgate
How to build a weighted criteria matrix
- List evaluation criteria that matter to the decision: effectiveness, cost, implementation time, risk, regulatory impact, sustainability of gains.
- Assign weights to each criterion (e.g., sum to 100%). Weight reflects relative importance.
- List solution options in columns.
- Score each solution against each criterion (e.g., 1–5 scale, where 5 = best).
- Calculate weighted score = score × weight for each cell.
- Sum the weighted scores for each solution. Highest total wins.
Worked example: Examplorphine fill weight solution selection
A. Re-calibrate auger control module to current spec
B. Install real-time gravimetric closed-loop fill weight control
C. Add in-process check-weigh station with automated reject
| Criterion | Weight | A: Recal. | A×W | B: Closed-loop | B×W | C: Check-weigh | C×W |
|---|---|---|---|---|---|---|---|
| Effectiveness (reduces fill weight error) | 30% | 3 | 0.9 | 5 | 1.5 | 4 | 1.2 |
| Implementation cost | 25% | 5 | 1.25 | 1 | 0.25 | 3 | 0.75 |
| Implementation time | 20% | 5 | 1.0 | 1 | 0.2 | 3 | 0.6 |
| Sustainability / recurrence prevention | 15% | 2 | 0.3 | 5 | 0.75 | 3 | 0.45 |
| Regulatory / validation risk | 10% | 4 | 0.4 | 2 | 0.2 | 4 | 0.4 |
| Total weighted score | 3.85 | 2.90 | 3.40 |
Recommendation: Solution A (recalibration) scores highest and should be implemented immediately. Solution C (check-weigh) is the second choice for longer-term defect detection. Solution B has the highest long-term effectiveness but high cost and validation risk for a GB-scope project.
Pugh concept selection (simplified variant)
In a simpler variant, each solution is rated +, —, or S (same as a reference/baseline) for each criterion, then +'s and −'s are summed. The solution with the most +'s relative to −'s is the leading concept. Useful early in the Improve phase when quantitative scoring data is not yet available.
Piloting solutions before full implementation
Before full rollout, run a pilot:
- Implement the solution on a small scale (one line, one shift, one operator)
- Collect data to confirm the improvement matches the prediction
- Document any unforeseen side effects
- Use the pilot results at the Improve tollgate to justify full implementation
Common exam traps
- Solutions are not identified in Define or Analyze. The charter goal and root causes are confirmed first; solutions come in Improve. A charter that says "install a vision system" has confused solution with goal.
- Weights must reflect team consensus. If one person assigns all the weights, the matrix reflects only their priorities. Use a team vote or AHP for high-stakes decisions.
- A high score does not automatically mean "implement." Check feasibility (budget, timeline, regulatory) before recommending the winner.
The free 20-question CSSGB diagnostic covers all six Body of Knowledge sections, with an explanation for every answer.
Take the free CSSGB diagnosticMore: Fishbone & 5 Whys · Poka-yoke · Control plan · 8 wastes · DMAIC overview
CSSGB Quest by Bizarre Act. Original study material; not affiliated with ASQ. CSSGB is a certification of ASQ. Terms · Privacy