KEY TAKEAWAYS
- Preserve original event intervals when adding or correcting reasons.
- Review stop frequency and duration together.
- Use comparable periods and connect the chart to a follow-up action.
In this resource
Keep events before grouping reasons
A downtime Pareto chart ranks categories so a team can decide what to investigate first. Its usefulness depends on the event model below it: the start, end, machine, state and recorded reason need to remain available for review.
An event initially recorded as unknown may receive a reason after an operator or technician reviews it. Preserve the original time interval and make the correction process clear. Otherwise the team cannot tell whether a changed total reflects a process improvement or a reporting edit.
Keep the reason hierarchy usable
Choose a manageable set of reasons that distinguish actions the team can take. Material waiting and equipment failure may require different owners; a long list of nearly identical electrical reasons may be hard for an operator to apply consistently.
Define whether setup and planned maintenance are in the analysis boundary. Keep offline or missing-data periods visible as a data-quality category where their operating state is unknown. Hiding those intervals makes the chart look more complete than the evidence supports.
Compare duration with frequency
A category with the most lost minutes may be driven by one unusual incident. Another category may contain many short interruptions that recur every shift. Look at frequency, total duration and event distribution together before deciding where to act.
For example, a single 90-minute stoppage and eighteen five-minute stops have the same total duration but may call for different investigations. These are illustrative values. Check individual timelines and the product or shift context before assigning a corrective action.
Compare like production periods
Review the number of scheduled hours and the set of machines included in each period. A busier week can accumulate more downtime minutes even while its loss rate improves. State the denominator if you normalise losses per planned production hour.
Product mix, staffing and changes to the reason hierarchy can also affect the comparison. Freeze reporting definitions for the assessment period or clearly mark a change. Avoid presenting a reordered chart alone as proof that a corrective action worked.
Close the loop with an owner and evidence
Select a small number of events or categories, assign an investigation owner and agree what evidence would demonstrate improvement. Recheck the cause with production and maintenance rather than assuming the selected reason is a confirmed root cause.
WhereFy Smart Insight can support the machine timeline and shift-reporting part of this workflow. The important handover is the connection between a recorded event, a review decision and a follow-up action; monitoring data alone does not eliminate downtime.
Have a site-specific question about these ideas?
Discuss your application