First Pass Yield Improvement: Coach's Guide

Welcome — why FPY matters to your team

First Pass Yield (FPY) measures the share of units that successfully complete a process step (or sequence of steps) without rework, repair, or scrap. Improving FPY lowers cost, reduces lead time, frees capacity, and builds confidence in the process. As a coach your role is to help teams detect defects earlier, simplify checks, and embed repeatable experiments and coaching so improvements stick.

Quick orientation

This guide helps you:

• Map where defects are discovered and their true cost.
• Shift inspection left and use simple poka-yoke (mistake-proofing).
• Run short, measurable A/B experiments to find what works.
• Coach consistently and avoid common measurement traps.

1. Prepare: establish a clear, reliable baseline

Before making changes, make sure FPY data is trustworthy and that the team agrees on definitions.

• Define the scope: Is FPY for a single station, a multi-step process, or end-to-end? Write the exact start and end points.
• Define a defect: What counts as an escape, rework, or scrap? Use clear, observable criteria (dimensions, function, cosmetic level) to avoid ambiguity.
• Collect baseline data: Track the number of units entering the scope, units requiring rework, and units scrapped over a short consistent period (e.g., one week or one shift pattern).
• FPY formula: FPY = (Units completed without rework) / (Units started in the scope). If you record rework events per unit, clarify whether you measure unit-level FPY or step-level FPY.

2. Map defect discovery points (value of knowing where defects are found)

Create a simple map that shows when defects are discovered relative to value-adding steps. The goal is to see how many defects are found late (after many steps and cost added).

1. List process steps in order.
2. Annotate where defects are discovered today and classify the defect types.
3. Estimate the relative cost or time to fix defects discovered at each point.

This map helps prioritize where shifting-left will have the biggest payback.

3. Shift-left inspection and poka-yoke ideas

Moving checks earlier prevents adding unnecessary value before detecting an issue. Prefer simple, low-cost fixes before complex automation.

• Stop the line triggers: Quick, discrete checks that let an operator stop and ask for help when something is out of tolerance.
• Visual cues: Color coding, go/no-go fixtures, labeled stations, or process windows that make incorrect states immediately obvious.
• Built-in physical constraints: Parts that only fit one way, keyed features, or fixture geometry to prevent incorrect assembly.
• Simple sensors: Microswitches, presence sensors, or low-cost inspection jigs placed earlier in the flow.
• Decision aids: Short checklists or two-step verifications embedded in the workstation or digital work instruction.

4. Run quick A/B experiments for inspection tactics

Use timeboxed experiments to learn fast. Keep experiments small, measurable, and reversible so they reduce risk and build confidence.

Experiment template (use on a whiteboard or as a simple form)

1. Hypothesis: If we add X (e.g., a go/no-go gauge at Operation 2), then defect type Y discovered at Operation 6 will fall by Z% within one week.
2. Baseline: current FPY for the defined scope and frequency of defect Y.
3. Intervention: Describe exactly what will change, who will do it, and when.
4. Duration: Timebox to 3–10 production shifts depending on volume.
5. Metrics: FPY in scope, defect counts by type, time-to-fix, and any unintended side effects (throughput change, cycle time impact).
6. Decision rule: Predefine what counts as success (e.g., 30% reduction in that defect with no negative impact on cycle time) and what to do if results are inconclusive.
7. Next steps: If successful, standardize; if not, iterate or rollback.

5. Coaching on the floor: practical scripts and habits

Coaching creates consistent practice and prevents improvements from fading.

• Before an experiment: Brief the team on the hypothesis, how to run the test, what to record, and emphasize learning over blame.
• During the shift: Use short huddles (5 minutes) at shift start and end. Watch 3–5 cycles with the operator, ask open questions, and note observations.
• When defects occur: Use a short 'stop, describe, tag' script: stop production if safety or major quality impact; describe what you saw; tag the part and capture photos or measurement data for root-cause analysis.
• End-of-test review: Share results visually, compare baseline vs. test, and decide next steps with operators involved.

6. Measurement hygiene and common pitfalls

• Avoid vanity metrics: Don't let a reported FPY number hide instability. Look at defect types, trends by shift and operator, and test sample sizes.
• Be clear on counting rules: When multiple defects occur on the same unit, record unit-level FPY consistently (unit failed or passed) or record per-defect counts — but don't mix the definitions.
• Watch data quality: If inspection moves locations, ensure the new inspection is consistent with previous methods; otherwise the apparent improvement may be an artifact.
• Balance speed and quality: Monitor cycle time and throughput during experiments to ensure a fix doesn’t create other problems.

7. Examples of small, high-impact experiments

• Install a tactile go/no-go gauge at a preparatory step to catch incorrect dimensions before assembly.
• Introduce a two-second visual check with a checklist before a process that adds value — compare FPY for three shifts before and after.
• Swap a flexible fixture for a keyed fixture on one machine and measure defects and changeover time over a week.

8. Coach's quick checklist

1. Agree measurement definitions with the team.
2. Map defect discovery points and estimate fix cost by location.
3. Run one small, timeboxed experiment this week with a clear hypothesis and metrics.
4. Observe operators, collect direct evidence (photos, parts, measurements), and avoid second-hand reports.
5. Standardize successful fixes into standard work and update training/work instructions.

9. Next steps and scaling

After local wins, package successful checks, fixtures, and experiments as standard work templates so other lines can trial them. Use regular cross-shift or cross-line huddles to spread learning and avoid isolated solutions.

10. Keep the team engaged

Celebrate small wins publicly, connect FPY improvements to customer impact and operator workload reduction, and keep experiments visible so the team owns the learning.

Closing note

This guide focuses on practical, reversible steps you can coach into everyday work. Real FPY improvement comes from repeated small experiments, strong measurement hygiene, and consistent coaching—not inspections that simply move failure detection downstream.