Production Scrap Reason Pareto

A factory records scrap, but a total waste number does not tell managers where to begin, so a Pareto view groups scrap by recorded reason and sorts the groups from largest to smallest to help choose which cause to investigate first. Define scrap by identifying material or units that cannot become acceptable output under the chosen boundary, with reworkable units and normal process trim needing separate labels.

Choose the measure, since count of rejected units, scrap weight and cost can rank reasons differently, so state the unit on the chart and avoid mixing measures. Agree on reason codes such as contamination, short fill, wrong dimension, damage and setup error, and make each code describe observable evidence rather than guess an untested root cause.

ASQ explains that Pareto bars can represent frequency or cost, ranked highest first, and the chart highlights significant categories but does not prove that the largest one is easy or safe to fix. Collect a consistent period, because a week may be useful for a busy line while low-volume production needs a longer window, and mark the date range and product scope.

Assign one clear category per scrap record when using a simple chart, and if multiple causes are tagged, define whether a single record contributes to several bars and explain the resulting total. Keep an other category limited, since a large unclassified bar prevents a useful priority decision, so review notes and improve the reason list if it grows.

If a plant scraps 50 units for short fill, 30 for contamination and 20 for damage, those bars appear in that order, and a cumulative line shows 50%, then 80%, then 100% of the 100 recorded scrap units. That example does not establish a universal 80/20 rule, since the first two causes happen to account for 80% in the invented set and real distributions may differ.

Check denominator and volume, because a reason share among scrap does not tell the overall scrap rate, and a line with 100 defects out of 100,000 units differs from one with 100 out of 500. Compare comparable processes and watch shifts and lots, since a cutting line and a filling line have different failure modes and a spike in one supplier batch or shift may point to a different cause than a stable plant-wide issue, so stratify before buying equipment.

Include cost if relevant, because a rare but expensive failure may deserve priority despite a short count bar, so use a second chart rather than changing units mid-chart, and validate data capture, since if workers are pressed to choose the quickest code the chart reflects convenience rather than production reality. Distinguish symptom from cause, as a cracked part can result from handling, temperature or material quality, so the Pareto category is a starting point for root-cause analysis, and talk to operators, whose insight should be tested against records.

Prioritise by impact and feasibility, because the largest category may require a long redesign while a smaller safety issue needs immediate action. Run a controlled improvement by changing a setting, tool or handling step with documented conditions, then measure later scrap and quality and compare fairly, and check unintended shifts, since reducing short fills by overfilling can raise material use, so monitor yield, quality and costs together.

Preserve the old chart, because a new coding scheme can make apparent improvement that is only relabelling, so annotate changes and compare on compatible definitions, and review trends since a cause that falls in one week may rebound. For owners, the Pareto view turns scattered reject records into a focused question, saying where most measured loss sits, not why it happens or which fix will work.