« Another Reason for Medical Mistakes - Fatigue | Main | Tech Support Thursday: PDF Printing For Free »
September 07, 2006
ALASKA PIPELINE FAILURE
More data has been released concerning the Alaskan oil pipeline leak that shut down a major portion of the Prudhoe Bay oil field on the Alaskan North Slope. It appears that BP changed their pipeline PM requirements based on the history of failures, then did not check to see if this new schedule was working correctly. After years of running a pig through the pipeline at fairly close intervals to clean the pipes, they decided to stop the cleaning and only conduct spot ultrasonic testing of the piping instead. Another pig inspection and cleaning was slated for next year (9 year interval), but leaks were found last month that required the shutdown. The company now plans to replace 16 miles of deficient piping.
Anytime a major change is made to a PM schedule, many risks must be considered. In hindsight, it may be easy to say that BP poorly anticipated the consequences of their change in maintenance strategy, but how do you mitigate these possible consequences?
It would be nice to know what the possible modes of failure are when changing (or initially developing) a maintenance plan. Once these failure modes are known, you can tailor your preventive maintenance to target the most likely (or catestrophic) failures.
Equifactor® to the rescue! When you initially install a new piece of gear, why not take a look at Equifactor® to determine how your machine might fail? Set up your PM schedules to target these failure modes, and get rid of those PM's that are not relevant to your piece of gear. This will allow you to funnel your maintenance dollars toward the areas actually needing the maintenance. Consider Equifactor® to be an important tool in your RCM toolbox.
Another consideration for developing (or changing) your maintenance strategy is the possible consequences of a failure. Drastic changes in a critical system may not be advisable. Smaller changes, or additional checks, might need to be instituted to catch costly failures resulting from changes to the maintenance plan.
Posted by kenreed at September 7, 2006 11:47 AM
Trackback Pings
TrackBack URL for this entry:
http://satyrs.site5.com/~taprootc/mt/mt-tb.cgi/182
Comments
Are there readily accessible sources to document the BP changes to their pipeline PM schedule?
If so, please could you post a couple of links.
Thanks
Posted by: Mike Baer at September 12, 2006 11:35 AM
Mike,
The full details are still being worked on. The info here was from an e-Newsletter I subscribe to, and really had no further data than what I stated here. I'll find the article and post a link.
The main point to get is that, while traditional RCM analyses may be used to adjust maintenance periodicities, much thought must be applied to your changes to ensure they are analyzed for unintended consequences. Changing the periodicity on a PM may make perfect sense in the analysis; however, processes that have a high risk of catastrophic damage may require additional verifications to rule out unintended consequences and insure the integrity of the system.
Also, Equifactor® can be used to look for some of these unintended consequences, as well as help in the initial design of the PM system for new equipment.
Ken
Posted by: Ken Reed at September 13, 2006 06:52 AM