No hunman and no organization is perfect! Although we should strive for perfection in all aspects of business (Production, Quality, Cost Control, and EHS), we often mis-step or possible loose our way altogether. But how bad do things have to get before it is a REAL PROBLEM?
Well…we like to tell our clients there is a BIG difference if OSHA/EPA finds before an accident than if they find it after an accident. A lot of plants have set a measurement for Work Orders based on 30-day past due, 60-day past due, and 90-day past due in order to have some type of gauge for work order completion. This is a good practice, as long as the plant knows that after an accident, all bets are off and ANY PAST DUE work orders will be scrutinized to the nth degree. And just like the Japanese Nuke plant news that surfaced last week that they missed key inspections on emergency generators that powered the emergency cooling water systems. We have no way of knowing if these missed inspections were a critical path in this incident; but to the tens of thousands of homeowners who have been pushed from their homes, not by the earthquake or the tsunami, but by the nuclear incident that followed, these inspections become a critical path through emotions. So the question that is always asked by management…how far is too far?
We like to use a weighted scale in our explianation. In other words, it depends on the device, the type of inspection, and the criticality of the device’s reliability. Of course this all plays into how big of an incident would result in the failure of this “safe guard”. Using your PHA, look back at the scenario(s) that claimed this equipment as a safeguard. If your PHA is in some software that calculates a risk factor for the scenario, play with the numbers by taking out the safeguard and seeing what level your risk rises to. Now I am not implying one missed inspection, or even two missed inspections will cause the incident; what I am trying to show is that if !@#$% hits the fan and these missed inspections will come to light and at least the business will have sound data to measure the seriousness of the missed inspections.
The better way to manage missed PMs is a regular review of PM’s. Sound Risk Based Inspection programs will routinely widen time periods between inspections that show a statistically valid pattern of PASSED inspections. In other words, a piece of equipment that has been on a 30 day inspection frequency can be extended to say 60 days, 90 days or even quarterly if the data supports this. This allows time and resources to be devoted to other critical areas. Of course this methodology has to be used on the other side of the equation…if the PM on a piece of equipment is ALWAYS FAILING then we must narrow the inspection frequency to ensure the equipment does not fail and cause an incident.
Bottom line, missed inspections may or may not play a role in an incident, but after the incident the whole world may be the judge and that is a court room you do not want to be sitting in! Manage your inspection frequencies. If you have solid data that frequencies can be extended, do an MOC and extend them; however, if you have inspections that show a pattern of problems, you need to analyze this and determine if the frequency needs to be shortened. Whatever the decision, it should be based on recognized and generally accepted good engineering practices (RAGAGEP) and documented in the equipment files.