Efficiency Metrics System —— OEE Decomposition, FPY, and Value Stream Efficiency Metrics

1. Efficiency Metrics: Not Just an OEE Pie Chart

At operational meetings, OEE 75% — no one can clearly explain whether it is low availability, low performance, or low quality rate; the quality department's PPM is decreasing, but per capita output and labor cost per unit are deteriorating.

The efficiency metrics system helps organizations break down waste, align improvement projects, and avoid misleading by a single metric. It is not in opposition to quality metrics: the quality rate is one dimension of OEE; rework affects both quality and efficiency.

2. Efficiency vs. Quality: Cause and Effect, Not Trade-offs

Suggestion: The value stream team should share responsibility for RTY/FPY + OEE + cycle time, with the quality department providing quality loss breakdown.

3. Hierarchical Efficiency Metrics in Manufacturing

3.1 Equipment Level (OEE Decomposition)

• Availability = Operating time / Planned production time (downtime, changeover, material shortage)
• Performance = Actual output / Theoretical output (minor stops, speed reduction)
• Quality = Good products / Total output (including startup scrap, consistency in strategy required)

OEE = A × P × Q; improvements must first identify the weakest dimension.

3.2 Value Stream Level

• Cycle Time, Lead Time
• WIP Turnover Days, Line-side Inventory Value
• Changeover Time (SMED), First Article Inspection (FAI) Pass Rate
• On-Time Delivery (OTD) — a comprehensive measure of planning and execution efficiency

3.3 Factory / Corporate Level

• Per Capita Output, Labor Cost per Unit
• Energy Consumption/Material Utilization (linked with Cost of Quality)
• Order Commitment Fulfillment Rate

4. Quality-Related Efficiency Metrics

Note: The numerator and denominator definitions should be consistent across the entire company (e.g., whether FPY includes rework after qualification).

5. Example of a Metrics Tree (Discrete Assembly)

OTD ↑
 ├─ Plan Achievement Rate ↑
 ├─ Material Shortage Incidents ↓
 └─ Manufacturing Lead Time ↓
      ├─ OEE ↑ (A/P/Q)
      ├─ WIP ↓
      └─ FPY ↑
           ├─ Incoming PPM ↓
           └─ Process Cpk ↑

Improvement projects should be linked to a specific node in the tree to avoid "improvement activities unrelated to KPIs."

6. Target Setting and Baselines

1. Collect 6-12 months of baseline data (avoid arbitrary decisions)
2. Set targets by customer/line (mixed product factories need stratification)
3. Define improvement targets (e.g., OEE +3pt annually, FPY +2pt)
4. Define the proportion of automatic data collection (manual entry can lead to inaccuracies)
5. Yellow/Red thresholds trigger QRQC

7. Kanban and Meeting Rhythms

Top 5 Losses (Pareto) should be in a fixed format: reason classification, duration, DRI, status.

8. Relationship with COQ and Lean Tools

• External Failure COQ ↓ often accompanies FPY ↑, OEE Q ↑
• SMED improves A dimension; minor stop improvements improve P dimension
• Kanban supermarket improves WIP and Lead Time
• Efficiency projects must assess quality risks (e.g., skipping steps to speed up)

The Quality Think Tank's articles on cost of quality management and lean logistics can be read in combination.

9. Common Misconceptions

Misconception 1: One OEE Number for the Entire World Multiple lines mixed together can mask issues — report by line and product family.

Misconception 2: Quality Rate Excludes Startup Scrap Inconsistent with financial and customer reporting — define in the quality manual.

Misconception 3: Efficiency Projects Do Not Validate Quality Complaints increase after speeding up — project closure requires 90-day FPY/complaint monitoring.

Misconception 4: Metrics Have No Owner Good or bad numbers can't be traced to anyone — clear RACI.

10. Implementation Checklist (Joint Quality/Operations)

1. Standardize OEE, FPY, Lead Time definition manuals
2. Select one pilot value stream, automatically collect MES data
3. Establish daily 15-minute loss review
4. Monthly Pareto update + improvement project pool
5. Management review input efficiency and quality trends

11. Data Collection and Systems

Project: Start with automated data collection for one value stream before expanding; define a data steward to ensure consistency.

12. Template for Linking Improvement Projects to Metrics

Each lean/Six Sigma project charter must include:

• Associated KPI tree node (e.g., OEE-A dimension, FPY)
• Baseline / Target / Verification Period
• Quality guardrail metrics (e.g., complaints must not increase during the project)

13. Shift-Level OEE Loss Codes (Example)

Unified coding facilitates Pareto analysis:

The quality department should participate in defining Q codes to ensure production reports A/P and Q losses.

14. Joint Reporting with COQ (4.3.1)

Management review should be on the same page: external failure COQ trend + customer PPM + OEE quality rate + FPY — four lines moving in the same direction indicate win-win in quality and efficiency; if COQ decreases while OEE Q increases, verify consistency in definitions.

15. Template for Pilot Value Stream Weekly Reports

• OEE (A/P/Q) and top 3 loss hours
• FPY / RTY compared to last week
• WIP value and completeness rate
• This week's improvement items and quality guardrail results

When benchmarking across multiple factories, standardize loss codes and OEE calculation boundaries (whether to include planned downtime, startup scrap) to avoid discouraging participation. The purpose of benchmarking is to share best practices, not to penalize through rankings.

16. Interface with TPM

Equipment availability (OEE-A) improvement projects should share TPM work order data; the quality department should focus on whether Q losses due to improper adjustments are included in TPM verification. During the ramp-up phase of new lines, do not set OEE targets too early; first stabilize FPY and loss code accuracy before pursuing performance rate. Incorporating per capita output and quality rate into the value stream manager's evaluation can reduce short-term behaviors of "sacrificing quality for output." Remove unused metrics during quarterly reviews to prevent kanban metric inflation.

The value of the efficiency metrics system is to make "where the slowdowns and losses are" visible, discussable, and improvable — not to gloss over with a single OEE number.

Before improvement, ask: Has the number of hours for the top reason in last month's OEE loss Pareto changed? Who is responsible?

Knowledge code: 4.3.3

Version: v20260702

Author: Quality Think Tank The Quality Think Tank is dedicated to providing systematic professional knowledge, methodologies, and practical tools to help quality management practitioners continuously enhance their company's quality capabilities.