I recently found myself leading a maintainability workshop at a food processing and packaging plant. My mandate was to guide a team of designers, maintenance leaders and reliability engineers through a critical review of their machinery design practices so that their critical production assets could be more easily maintained and reliably operated. Part of this exercise was to take the group out into the plant and focus on one bottleneck area to ascertain what opportunities we could find for improvements. Although I was confident that we would find lots of room for improvement, nobody could have prepared me for what I was about to experience!
Packaging Line
While the plant or company shall remain nameless, our machine of choice was a packaging line. This line was one of several in the plant where finished product was stacked into cardboard boxes before proceeding to warehousing and distribution. Needless to say that when this machine goes down, the plant loses the equivalent amount of production handled by this machine. With little to no inventorying of product, even a minute of downtime represents a real cost to the organization.
For those not familiar with packaging equipment, this machine is a complex mixture of mechanical drives, chains and pneumatics. Oftentimes, many of the systems that require periodic maintenance are hidden behind safety guards such that any entry into the machines requires tripping an interlock at a minimum and, in some cases, requiring full lockout/tagout protocol.
As usual, I suggested to the group that we should engage the machine operator in our fact-finding mission. After all, most operators "live" with their equipment for a full eight-, 10- or 12-hour shift, meaning they have the best insight into what problems occur most often. In this instance, our operator was a 30-year veteran of the plant, who was just a few years away from retirement. For argument's sake, let's call him Joe.
Meet the Expert
In this plant, Joe had deservedly earned himself a reputation as being a top-notch operator. He knew the machines inside and out, and could get the packaging line back up and running faster than any other operator. In fact, the production manager for this line wished he had another half-dozen employees like Joe and was fearful of what would happen when Joe retired. True to form, Joe could not have been more accommodating, and he was more than willing to share his knowledge with us.
After explaining our mission to Joe, our team set about picking his brains to elicit what issues he had experienced with this particular machine. What a wealth of information Joe turned out to be! You see, during the hour or so we studied this machine, our team witnessed Joe responding to an operational issue no less than 40 times! While the reasons appeared to be many and varied, most centered on the box assembly machine, designed to take a box in flat form and assemble it into the container in which the product would be shipped.
This machine, like the rest, appeared to be a complex mixture of drive systems, chains, etc.; but the heart of the machine appeared to be the pneumatic system used to operate a series of suction cups used to assemble the box into its three-dimensional form. On at least 20 different occasions, I witnessed Joe open the machine to clear a jam, in each case tripping the machine interlock and halting production. Even though Joe was skilled and quickly resolved the issue, with a time per incident of approximately 30 seconds, it was clear that those 10 minutes of the hour I observed was lost production due to this one recurrent problem.
Pneumatic Challenges
Questioning Joe further, it became clear that pneumatic issues were the bane of Joe's life. Aside from jams caused by missassembly (something Joe attributed to the storage of the batches of new boxes in a humid production environment), he explained that he routinely changed the pneumatic hoses at least once per day on smaller cylinders, and once a week on larger lines. Looking closer, it was clear that the amount of flex, caused by tight curves, the bundling of hoses together such that they were rubbing on each other during each production cycle and the cheap inferior replacement hose (to try to minimize the high cost of replacements) were leading to premature failures of the hosing. In fact, Joe had dealt with this issue so often that he proudly stated he could replace a hose in less than three minutes!
At the end of the in-plant machine survey, the group returned to the classroom to join with other teams who had surveyed other machines. We talked about Joe, and a look of sheer amazement came across the faces of several participants. Nobody had even the slightest clue what was happening on an hour-to-hour, day-to-day basis. They knew that the machine was a "bad actor", but lacking specific data to ascertain the reasons why, they were simply relying upon Joe and his experience to overcome a serious reliability issue.
So what's the solution? Like most maintenance and reliability problems, what was lacking was data. Not macroscopic data such as overall equipment effectiveness (OEE), which the plant was tracking, but detailed information allowing each failure to be categorized according to a failure code (for example, a broken pneumatic hose) so that data could be analyzed for trends. Ideally, each failure code should be linked to a cost per minute or second so that the total cost of each failure mode can be determined, allowing management insight into its major cost areas. This process and the tools to capture the data are sometimes referred to as the failure reporting and corrective action system (FRACAS).
The Value of Failure Reporting
The value of FRACAS data cannot be overstated. While incidents that incur major time usually hit our radar screen, small problems that take a few minutes or less to fix but occur multiple times during the day can often be a larger contributor to overall downtime costs. This is a process analogous to death by a thousand paper cuts.
So if your plant involves a process where small incremental steps are involved as part of production, consider the value that FRACAS can offer in providing data - not just about major events, but those seemingly small, inconsequential events, because they add up!
As always, this is my opinion. I'm interested in yours.
