We’ve all been there – working through a Deviation or Complaint or Audit Finding, and then the question of the Root Cause comes up. You need the Root Cause in order to be able to Correct and Prevent the issue again in the future.
How many such analyses find that something happened due to a mistake being made; and that the most generalised reason is not that the people involved were useless at their job, but that the training was insufficient?
Well what if it were not either of these things – but external and higher-level factors, harder to grasp, that caused a mistake to be made? How do you trace those aspects, and fix them, rather than constantly sending people to be trained again, and improving the training, in an endless cycle?
Take these examples from our customer base:
- Over the period of 3 months, one customer logged 12 “human error” deviations in one of their production facilities. They were all initially classified as a training issue.
- Another customer had an ongoing set of human errors from a packaging facility, where there were multiple packaging assembly lines across the top floor of the facility. 90% of them were put down to human error / training (a couple of them were due to equipment failure).
Both customers have CARA implemented for their QMS and quality documents, and the first company mentioned above has been using CARA in their HR department for various documents including timesheets.
The first customer decided to use the HR timesheet data to analyse against the deviations, and with a single click report it was identified that all the staff involved in the deviations were working the night shift. The company decided this might be the real root cause, and instituted more frequent breaks, and the level of deviations dropped significantly over the following 3-month period.
The second customer used the geotagging and maps-based view of the deviations, and what stood out from there (which had not been apparent in dry columns of information and data) was that all the deviations were clustered at one end of the production floor – where, due to overhead duct work, the lighting was not as strong as in the rest of the facility. The customer guessed that this might be the true trigger of the deviations and has corrected the lighting; deviation levels are still being monitored but early results look like this did indeed play a large part in the level of issues.
This is not to say, of course, that no deviations are ever caused by a lack of proper or frequent training; but that it is not always the real reason. If you arm yourself with tools that can analyse and display / match data patterns at a higher level, you just might surprise yourself as to the causes of deviations and save your company a lot of money!
– James Kelleher, CEO, Generis