Rework level

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