OperationsFramework / modelModelAccessible

Rework level

How can rework level improve people, teams, or organisational effectiveness?

AccessibleOperationalIndividual3 min read

Contents

Helps managers answer: How effectively are we driving waste out of our processes?

Rework level measures how often an output must be corrected after it fails to meet its specification. It is a direct signal of process instability and a contributor to the cost of poor quality. The managerial objective is not to inspect defects into compliance, but to prevent them and detect causes as close to their source as possible.

When to use it

Answer the key performance question: “How effectively are we driving waste out of our processes?”

Include the KPI in the operational processes and supply-chain perspective.

Track first-time quality, correction effort and the customer consequences of non-conformance.

Use it to direct process improvement and learning, not individual blame.

Origins

The measure belongs to the traditions of statistical quality control, total quality management and lean production. Western interest grew sharply in the 1980s as manufacturers studied Japanese quality systems and the cost of defects. Philip Crosby’s zero-defects philosophy and Quality Is Free reinforced the principle that prevention and conformance cost less than failure and correction.

What it is

Perspective: Operational processes and supply chain perspective.

Key performance question: How effectively are we driving waste out of our processes?

Rework is additional effort applied to a non-conforming output so that it meets the required specification. It differs from scrap, concession, warranty repair and ordinary iteration that was intentionally designed into a learning process.

A rework rate indicates how often the process fails to produce an acceptable result first time. Cost of rework adds labour, material, delay, retesting, administration and downstream effects. Both views matter: a rare defect can be economically or medically severe, while frequent minor correction can consume large capacity.

Quality should be built into process design and controlled near the source. Early detection prevents additional work from being added to an already defective output. Patterns may identify unclear requirements, poor design, supplier variation, equipment problems, workload or training needs.

How to use it

Measurement

Write an operational definition of a completed unit, inspection opportunity, non-conformance and rework. Separate planned revision from avoidable correction. Stratify by process, defect type and severity while protecting enough volume to see systemic patterns.

Data collection method

Manufacturing systems may capture non-conformance and routing automatically. Service organisations can use case systems, reopened work, corrected documents or customer-reported failure. Validate classifications and make reporting psychologically safe.

Formula

Rework rate = reworked outputs / inspected outputs. If 100 products are inspected and 4 require successful rework, the rate is 4%.

Also calculate rework hours and cost when units differ substantially in effort or consequence.

Frequency

Match review to process speed and risk. High-volume manufacturing may monitor continuously or weekly; many service processes monthly; lower-volume settings monthly or quarterly. Escalate severe defects immediately regardless of cadence.

Source of the data

Use the production or service process, inspection and test records, non-conformance systems, time records and customer feedback.

Cost/effort in collecting the data

Automated traceability can require material investment, while manual systems are cheaper but prone to underreporting. Evaluate the system against avoided failure cost and improved learning.

Target setting/benchmarks

Zero avoidable rework is the aspiration, but intermediate targets should reflect baseline capability, measurement quality and risk. Never discourage reporting to make the number appear closer to zero.

Example

The following fictional case from Carpenter Group LLC illustrates how a small defect can propagate through the cost of poor quality.

Strategically Fine Writing Instruments employed 20 people and sold a pen for $89.95 retail and $42.25 wholesale. Production cost was $18.20, leaving $24.05 profit per pen.

In February 2006, a major customer returned 4,000 unsold leaking pens and reported having sold 2,500 during the preceding months. Return shipping cost $35.85. SFWI quarantined stock and reinspected it for $400. Investigation traced cracks to a refill-wall change introduced in October 2005.

The firm treated all post-change refills as suspect. Appraising 25,000 refills and 8,000 completed pens cost $2,500.

Registration data identified 52,952 owners from November 2005 through February 2006. SFWI also notified 664 retailers at $2.17 per letter. Returns from those retailers cost $14,608.00 and added 40,000 pens.

Rework took 5 minutes per pen. At a loaded labour rate of $15/hr, labour cost $1.25 per unit. The 8,000 completed pens required rework, and replacement refills cost $2.00 each.

Testing proceeded at two pens per hour with 15 seconds of test time per pen. The case states 126,664 total pens @12 per hour = 21,110 inspections at $0.125 each. The 25,000 refills in stock were scrapped.

Redesign cost $2,000.00 in engineering and an expedited replacement order cost $10,000. Warranty work added $1.25 labour, $2.00 for the refill and $1.00 per pen, totalling $4.25 per unit.

Inbound shipping from wholesale customers cost $14,643.85. Recall notices at $2.17 went to the 52,952 registered owners and 664 retailers.

The case calculated total poor-quality cost of $603,558.57. The exact arithmetic should be independently reconciled before reuse; the strategic lesson is the breadth of downstream consequences.

Top practical tip

Pair the rate with a defect Pareto, cost and time-to-detection. Investigate the process conditions that allowed the failure and verify that corrective action prevents recurrence rather than merely accelerating correction.

Top pitfall

Do not create a blame culture or targets that reward hiding defects. If reporting rework threatens an employee, the metric will improve while quality deteriorates. Protect reporting, examine system causes and distinguish human error from reckless or deliberate conduct.