Manufacturing & custom software · Bengaluru · Whitefish, USA
Manufacturing

How to Reduce Rejection Rate in Manufacturing: A Practical Step-by-Step Approach

A practical method to reduce rejection and rework in manufacturing: measure rejection properly, Pareto by defect, machine and product, find root causes, error-proof the process, control start-up and changeover, manage suppliers, use SPC, and sustain the gains with data.

By Corposol Team··5 min read
Close-up of a caliper measurement during inspection
Photo: Calipers in physics lab by User:Falcorian, CC BY-SA 4.0 (cropped)

Every rejected part costs material, machine time, labour and energy — and every part that reaches a customer defective costs far more in sorting, penalties and reputation. Reducing rejection is one of the fastest ways to improve a factory's profit without new equipment. This guide describes a practical, repeatable approach.

Step 1: Measure rejection properly

Many factories know their rejection only roughly, from monthly scrap weight or shift registers. To improve, you need:

  • Rejects by defect type — a short, agreed list of defect codes per process.
  • By machine, product, operation, shift and lot.
  • Rework separately from scrap.
  • Where it was detected — in-process, final inspection, or at the customer.
  • Cost — material and processing cost up to the point of rejection.

Record rejects at the machine when they occur, on a tablet or with a simple form, rather than at the end of the shift. See in-process quality checks.

Step 2: Find the biggest problems

Use a Pareto chart: rank defects by quantity or cost. Usually a few defects on a few parts and machines account for most rejection. Drill down: which machine, which shift, which material lot, which die or mould? Patterns point to causes.

Step 3: Contain, then find the root cause

Protect customers first. Then investigate the top defect with structured tools: 5 Why, fishbone diagrams, and comparison of good and bad parts. Consider the 5M: man, machine, method, material, measurement. Look at process data and maintenance history for the time the defect appeared. Manage the actions through CAPA.

Step 4: Fix the common causes

Machine condition

Worn spindles, guides, tooling and dies are frequent causes. Preventive maintenance and tool life limits prevent many defects. See tool life management.

Setup and changeover

Many defects occur in the first parts after a setup. Standard setup sheets, stored parameters and first-piece approval reduce them. These are the start-up rejects among the six big losses.

Method and instructions

Unclear or outdated instructions lead to variation between operators and shifts. Digital work instructions show the current revision, with photos of correct and incorrect results.

Material

Supplier variation in dimensions, hardness or composition. Incoming inspection, supplier corrective actions and supplier scorecards address it.

Measurement

Gauges out of calibration or inconsistent measurement methods produce false accepts and false rejects. Calibration control and measurement system analysis help.

Step 5: Error-proof the process

Poka-yoke prevents errors or detects them immediately: fixtures that accept parts only in the correct orientation, sensors that confirm a hole was drilled, counters that confirm all fasteners were used. Error-proofing is more reliable than inspection or instructions.

Step 6: Control the process with data

Statistical process control on critical characteristics shows when a process is drifting before it produces defects. Control plans define what is checked, how often and what to do when out of limits — see control plans.

Step 7: Sustain

  • Display rejection by line and defect daily on the shop floor.
  • Review the top defects in a short daily meeting.
  • Track CAPAs to effectiveness.
  • Update PFMEAs and control plans with lessons learned.
  • Recognise teams that reduce defects.

Measures to track

  • Internal rejection % or PPM, by line and product.
  • First-pass yield.
  • Customer PPM and number of complaints.
  • Cost of poor quality — see cost of quality.
  • Open CAPAs and their age.

How software helps

A quality management system connected to production records rejects at the source, links them to machine, tool, material lot and operator, builds Pareto charts automatically, and tracks NCRs and CAPAs. With traceability, a customer complaint can be traced to its cause and its scope in minutes. Corposol builds these systems for manufacturers — see quality management software development.

A worked example: reducing rejection on a machining line

A machining line producing shafts had a rejection rate that management wanted to halve. Rejects were recorded at shift end as a total. The first step was recording rejects at each operation by defect code on a tablet. Within three weeks, a Pareto showed that two defects — undersize diameter at the grinding operation and chatter marks at turning — accounted for most rejects.

Investigation found that grinding undersize occurred mostly in the first hour after start-up, as the machine warmed up, and that the operator compensated manually with inconsistent results. A warm-up cycle and a defined first-part check solved most of it. Chatter marks were linked to a worn tool holder on one turning machine and to insert changes being stretched beyond the recommended interval. A tool-life limit and a scheduled holder replacement solved those. With both actions in place, rejection fell substantially within two months — and the data made it possible to prove it.

Quick wins to try first

  • Record rejects by defect at the point of occurrence, starting today.
  • Make first-piece approval mandatory after every setup.
  • Check gauges in use for calibration status.
  • Review tool-change and die-maintenance intervals against defect patterns.
  • Put photos of good and bad parts at each workstation.
  • Hold a ten-minute daily review of the top defect.

Frequently asked questions

How is rejection rate calculated?

Rejected quantity divided by total quantity produced (or inspected) for the period, expressed as a percentage or in parts per million (PPM). Track internal rejection and customer rejection separately, and include rework as its own figure.

Should rework count as rejection?

Track it separately. Rework is not scrap, but it consumes time and capacity and often hides process problems. Many factories report first-pass yield — parts good the first time — to capture both.

What is a realistic target?

It depends on the process and starting point. A common approach is to halve the top defect within a quarter, then move to the next. Customer PPM targets are often much stricter than internal ones.

Can software reduce rejection?

Software does not fix the process, but it makes defects visible by machine, product, shift and cause, links them to process data and traceability, and tracks corrective actions — which is what drives improvement.

  • Reduce rejection rate
  • Scrap reduction
  • Rework
  • Quality improvement
  • PPM

Let's discuss your factory's software

Tell us about your process or idea. You'll get a free consultation and a clear, fixed quote — no obligation.

  • Free consultation
  • Prototype before you commit
  • Fixed, itemised quote
  • You own the software

Get a free quote

We reply within one working day. Your details are never shared.