Blog · Board and management reporting · Manufacturing
A Power BI quality dashboard compares first-pass yield, scrap, rework and the cost of quality losses. Follow a worked manufacturing example and the checks that tie every measure to inspection and cost exports.
A Power BI quality dashboard should compare first-pass yield, scrap, rework and their costs on the same production population. Its central check is simple: every unit has one first-pass disposition, and every dollar has a supporting cost row. A low scrap count alone does not prove that a plant has a low quality loss cost.
Microsoft describes a Power BI dashboard as a single page of highlights, with related reports supplying detail. The business specification below defines which highlights belong together before anyone chooses a visual. Covirage is not affiliated with Microsoft or Power BI.
Start with completed first-pass inspections for a closed production period. Keep a production lot identifier, product, plant, completion date, units entering the process and initial disposition. A separate rework export supplies hours and approved rates; scrap transactions supply quantities and valuation.
Use unique physical units for yield. An individual unit can have several defect codes, so a count of defect observations is a different measure. Retesting a unit must not create another unit in the production denominator.
For this example, initial dispositions are mutually exclusive: first-pass good, sent to rework, or immediately scrapped. Any units still awaiting an initial decision would need a separate pending category. The broader quality assurance dashboard covers the operational review around these categories.
First-pass yield = First-pass good units / Units entering the process
Recorded quality loss cost = Scrap cost + Incremental rework cost
Loss cost per unit started = Recorded quality loss cost / Units entering the process
The distinction between first-pass good output and output requiring repair also matters to equipment effectiveness: the OEE quality definition excludes parts that need rework. Do not replace first-pass yield with eventual acceptance after repair.
An illustrative Ohio components manufacturer reviews production completed during September 2026. Both plants have closed initial inspections for the listed lots. The plants make different products, so the table identifies the product as well as the plant.
| Plant and product | Units started | First-pass good | Routed to rework | Scrapped | First-pass yield |
|---|---|---|---|---|---|
| Dayton, housing A | 10,000 | 9,400 | 400 | 200 | 94.00% |
| Toledo, bracket B | 5,000 | 4,600 | 300 | 100 | 92.00% |
| Total | 15,000 | 14,000 | 700 | 300 | 93.33% |
Dayton's yield is 9,400 / 10,000 = 94.00%. Toledo's is 4,600 / 5,000 = 92.00%. The combined figure uses the totals: 14,000 / 15,000 = 93.33%, rounded to two decimal places. Preserve full precision in calculations and round only the displayed result.
The total disposition check is 14,000 + 700 + 300 = 15,000 units. The same check must hold on each plant row. This proves that the initial population is completely assigned; it does not prove that the rework lots have finished or that inspections were performed correctly.
The following cost assumptions are explicit inputs. Scrap is valued at consumed standard cost with no recovery proceeds. Rework cost includes additional labor only, at $30 per hour; no additional materials were recorded. These are recorded internal failure costs, rather than a complete cost-of-quality statement covering prevention, appraisal and customer failures.
| Plant and product | Scrapped units | Scrap unit cost (USD) | Scrap cost (USD) | Rework hours | Labor rate (USD/hour) | Rework cost (USD) | Total loss cost (USD) |
|---|---|---|---|---|---|---|---|
| Dayton, housing A | 200 | 12 | 2,400 | 80 | 30 | 2,400 | 4,800 |
| Toledo, bracket B | 100 | 20 | 2,000 | 90 | 30 | 2,700 | 4,700 |
| Total | 300 | 4,400 | 170 | 5,100 | 9,500 |
Dayton's scrap cost is 200 × 12 = $2,400, and its rework labor is 80 × 30 = $2,400. Toledo's corresponding calculations are 100 × 20 = $2,000 and 90 × 30 = $2,700. The combined loss is 4,400 + 5,100 = $9,500, or 9,500 / 15,000 = $0.6333 per unit started, rounded to four decimals.
Toledo's dollar loss nearly matches Dayton's despite its smaller production population. That gives the quality manager a place to investigate. It does not establish a cause, since bracket B and housing A have different unit values and production requirements.
Keep every quantity, yield, rework hour and labor rate unchanged. Change only Dayton's scrap valuation from $12 to $14 per unit because the approved cost basis has changed.
| Measure | Original | Changed valuation |
|---|---|---|
| Dayton scrap cost (USD) | 200 × 12 = 2,400 | 200 × 14 = 2,800 |
| Company scrap cost (USD) | 4,400 | 4,800 |
| Company rework cost (USD) | 5,100 | 5,100 |
| Company loss cost (USD) | 9,500 | 9,900 |
| Loss cost per unit started (USD) | 0.6333 | 0.6600 |
| Company first-pass yield | 93.33% | 93.33% |
The loss cost rises by 200 × (14 − 12) = $400 while physical performance stays unchanged. A dashboard that shows only dollars could suggest a process deterioration. Showing counts and cost assumptions beside the cost total identifies the valuation effect immediately.
Use a headline area for the inspection population, first-pass yield and recorded loss dollars. Under it, place the plant-and-product table and a breakdown of scrap versus rework. Include period dates and the valuation policy beside the figures so the reader can establish their scope.
The plant quality manager owns disposition accuracy and defect investigation. Production owns the run and lot counts; finance owns the approved cost basis and the reconciliation to expense or inventory adjustments. Agree who resolves differences before a monthly review.
For visual organization beyond this specific question, see Power BI dashboard examples. Measurement completeness is a separate issue from process quality; Power BI data quality dashboard examples address missing, duplicated or inconsistent records.
Before release, tie started units to the production export, dispositions to inspection records, scrap values to approved adjustments and labor hours to the rework register. Explain unmatched rows before treating the result as a performance comparison.
Bring an inspection export, scrap cost rows and rework records with plant, product and period identifiers. Covirage maps the columns, and its tools compute and check the figures; the external AI model explains the differences with every figure traced to its rows. Create a free Covirage account and bring a sample of your own data.
See cost variance analysis for the supporting approach to separating quantity and cost effects.
Start with units entering the process, first-pass good units, units routed to rework, scrapped units and their costs. Keep customer returns separate from production inspection results because their dates and populations differ. Add an owner and reporting period to each measure.
Units that require rework do not count as first-pass good units, even if they later pass inspection. Record their eventual disposition separately. Otherwise a successful repair hides the original process failure and makes first-pass yield look better than the underlying production run.
Use the same valuation policy, reporting period and cost categories at each plant. Show the dollar total alongside a measure such as loss cost per unit started. Compare similar products before attributing a difference to plant performance, since product costs and process requirements can vary.