Dover PMAR 2010 - Maintainability

Editor's Notes

• #3: (c) 2007 Jerry Dover, P.Eng. "Maintain-ability": My focus will be on improving maintainability on INSTALLED, IN-SERVICE equipment This is a critical, basic maintenance process along with skills training, cost management, corrective and preventive maintenance (CM and PM), accurate lubrication, etc.
• #4: (c) 2007 Jerry Dover, P.Eng. NOT Reliability – rather it determines how well we respond to poor reliability. The total life of a system consists basically of MTBF + MTTR.
• #5: (c) 2007 Jerry Dover, P.Eng. Machine Characteristic – determined by design and influenced by maintenance department capabilities Serviceability: how easy/difficult is it to perform servicing (inpsections, lubrication, adjustments) required to maintain desired performance Reparability: how easy/difficult is it to perform replacement or restoration work in the event of a failure
• #6: (c) 2007 Jerry Dover, P.Eng. Improve maintainability Improve ability to service running equipment: 5-S, Poka Yoke, TPM, Lubrication Techniques, RCA, FMECA What remains surprising, though, is how little of it we really do. Today I want to advocate a relentless, fanatical approach to maintain-ability. The primary benefits will go to your business' bottom line, and your maintainers will also benefit along the way.
• #7: (c) 2007 Jerry Dover, P.Eng. Most maintenance professionals will agree that maintainability is best affected early in the design stage of a piece of equipment Lots of Giants in our field have written extensively on Reliability and Maintainability in Design: Dr. B.S. Dhillon at the University of Ottawa (Engineering Maintainability, Gulf Professional Publishing, 1999) B.S. Blanchard et al (Maintainability, John Wiley & Sons, New York, 1995) AKS Jardine et al (Maintenance, Replacement, and Reliability: Theory and Applications, CRC Press – Taylor & Francis Group, Boca Raton, FL, 2006) Organizations such as the SAE, Army, and others The reality however is that very few of us get to actually influence the design at this stage
• #8: (c) 2007 Jerry Dover, P.Eng. Our challenge is that for all new equipment being designed, there is a massive inventory of installed production equipment upon which companies must continue to rely
• #9: (c) 2007 Jerry Dover, P.Eng. There are many compelling reasons to focus on improving the reliability and maintainability of the equipment which you already have in service. maintenance and repair costs are frequently a significant portion of a plant’s operating cost Reductions in maintenance spending are immediate bottom-line gains
• #10: (c) 2007 Jerry Dover, P.Eng. Two big demographic factors are also putting pressure on manufacturing: The aging workforce, especially challenging with skilled trades due to the massive loss of experience that we forsee The very low enrolment in trades training, almost a full generation-old problem These two factors add up to indicate that there simply will not be the same quantity of skilled labour to carry out repair and maintenance tasks as we have done in the past. We must find new ways to do more with less, or DO LESS. I would like to describe to you a real example of maintainability improvement, see what lessons were learned, and discuss the basic principles which we can re-apply to other challenges. While it’s traditional to use an icebreaker or tell a joke to start a presentation, I’m too nervous for that so instead I’m going to tell you …
• #11: (c) 2007 Jerry Dover, P.Eng. When equipment or a process is scheduled and desired to run, it may meet performance goals or it may fail to meet those goals. The key is to define what is meant by “up”, or running to desired performance.
• #12: (c) 2007 Jerry Dover, P.Eng. Normally expressed in hours of operation, and calculated over a time frame long enough to be representative of typical lifespan behaviour of the equipment. Note that this number may not be valid during lifetime extremes, such as during burn-in periods, or wear-out near the end of the equipment’s service life.
• #13: (c) 2007 Jerry Dover, P.Eng. Also expressed in hours of operation, again calculated over a time frame long enough to be representative of typical lifespan behaviour of the equipment. Note that if the MTBF is quite large (i.e. reliability is high), then there may be little desire to improve MTTR. It is when reliability is LOW, and not easy to improve, that effort should be expended on maintainability.
• #14: (c) 2007 Jerry Dover, P.Eng. A lot of components over and above the wrench time: Identify: need to recognize that performance is outside of desired parameters and maintainers must intervene, and determine where the repair or replacement is needed, i.e. troubleshooting Remove from Service: prepare the machine or process for the maintenance intervention, especially important in continuous processes, confined spaces, hazardous environments, etc. Plan/Prepare: determine who-what-where-when-how for: trades skills, parts needed, location of spares, special tools or fixtures, time and method of access, safety precautions Execute Repair: actual “wrench time” to effect the repair or replacement Test/Return to Service: prepare to bring the machine or process back into normal operations
• #15: (c) 2007 Jerry Dover, P.Eng. … a Horror Story. What’s scary about it is that it is real, and not uncommon! Many years ago, I made a career move, from the army into food manufacturing. From maintenance platoon commander of a mechanized infantry battalion, to a maintenance supervisor in a unionized food plant. Simple move, easy as pie, nothing I couldn’t handle, right?
• #16: (c) 2007 Jerry Dover, P.Eng. A classic Bottleneck, which unfortunately was installed less than a year before, at great expense. Within my first month on the job, we had major breakdown at this bottleneck. As my maintainers struggled for almost an hour to get it running, the entire plant eventually ground to a halt, and the plant manager got more and more creative in expressing his opinion on the matter. It would be over two hours before we finally got things running again. Unfortunately, this breakdown at the bottleneck was not a one-off. It actually was happening every few months! This is high MTBF, or low reliability, with a high cost or consequence of failure.
• #17: (c) 2007 Jerry Dover, P.Eng. The bottleneck machine stopped. Product started to accumulate, as the mechanics responded and tried to get it running again.
• #18: (c) 2007 Jerry Dover, P.Eng. Eventually production lines started shutting down, idling a lot of people and processes. The mechanic couldn’t figure out what to do, and so called an electrician to help.
• #19: (c) 2007 Jerry Dover, P.Eng. The electrician couldn’t find any drawings, and was not trained on the machine. The lockout dropped all power from the machine, including control power, so there was no easy way to test and adjust. I was unable to give management a time estimate for the repair, so they could not plan what to do with all of the production people.
• #20: (c) 2007 Jerry Dover, P.Eng. We eventually had 4-5 guys trying to work on it. At some point during the troubleshooting, someone mentioned there was a SPARE strapper-labeler! Of course, I lost my top – what are we doing fighting with the broken one? Then another learning: it took another hour to get the spare installed … and then another learning: it didn’t work!
• #21: (c) 2007 Jerry Dover, P.Eng. Once the spare was installed, it was a slow process of testing and adjusting it before getting back to regular production.
• #22: (c) 2007 Jerry Dover, P.Eng. This breakdown was finally resolved, with over two hours of effort! Problems: No strategy or repair plan, so maintainers forgot the “big picture” of production – tunnel vision It took 2-3 people to remove and re-install the failed component No spare frame or jig – had to drop the failed component on the ground Needed to mark the wires to reconnect the spare properly In a unionized environment, several trades were needed to perform one replacement job
• #23: (c) 2007 Jerry Dover, P.Eng. Over two hours of downtime! The plant had endured this situation almost monthly for over a year. Now, I was still so fresh out of uniform that I would snap to attention whenever my boss walked by, so my natural instinct was to … ATTACK THE PROBLEM!
• #24: (c) 2007 Jerry Dover, P.Eng. Actually, there’s a good lesson to be learned from tanks. In the Canadian Army, the standard for replacing the engine and transmission in a Leopard 1A7 tank in the field is 30 minutes. This is done by the 4-man crew and two maintainers, day or night, in any season. This was the challenge I gave to my crew: modify the strapper to cut the replacement time to 30 minutes or less.
• #25: (c) 2007 Jerry Dover, P.Eng. Visual Indicators: gauges (with ideal and warning settings MARKED), stack lights, LEDs, coloured wear strips, etc. Use of Human-Machine Interfaces: not just alarm messages, program screens to show input and output states to avoid needing to open panels, find drawings or plug in laptops just to check I/O.
• #26: (c) 2007 Jerry Dover, P.Eng. A maintenance plan MUST include guidelines around how much troubleshooting time to tolerate at bottlenecks, before bringing a replacement on line. Remember that maintainers (including supervisors) are prone to tunnel vision! Don’t be afraid to modify equipment that must be repaired quickly. Care must be taken to avoid adding more failure modes.
• #27: (c) 2007 Jerry Dover, P.Eng. Don’t underestimate the power of DRILLS. Just like a sport, or emergency response, if it’s a critical action then you must practise it.
• #28: (c) 2007 Jerry Dover, P.Eng. More planning and preparing means less wrench time is needed. Preparation avoids wasting time hunting for tools or parts, looking for welding outlets, waiting for specialist trades, etc. As the stress builds, leaders need to insulate their people from organizational pressure: no one needs several people looking over their shoulder! If you’ve prepared and trained for it, then don’t run! It only panics the troops.
• #29: (c) 2007 Jerry Dover, P.Eng. Intend to fix it right the first time. Stay in contact with the equipment until it is running to specifications again. The Time To Repair is not over until quality product is being produced again. If there are repairable parts, then the tagging/assessment of them for rebuild MUST be an integral part of the process.
• #30: (c) 2007 Jerry Dover, P.Eng. Design: Bottleneck, hard-wired, supported on the frame by mounting bolts, Tools: Needed several different wrenches, screwdrivers Procedures: No standard – maintainers get tunnel-vision – and no rebuild procedure for the spare Skills: lack of training, troubleshooting, no Pit Stop mentality
• #31: (c) 2007 Jerry Dover, P.Eng.
• #32: (c) 2007 Jerry Dover, P.Eng. There were five essential changes. NOT RELIABILITY, BUT MAINTAINABILITY CHANGES: Preparation: we made a wheeled stand that holds TWO units back-to-back to reduce manual handling Mechanical: we bolted rails to the frame, and rest the machine on them, with hand clamps to secure it Electrical: we installed military-style (HUA!) quick disconnects Training: all maintainers were trained on the procedure for swapping out the machine Procedures: we instituted a rebuild procedure, so any swap like this triggers a repair work order – the job is not done until we’re ready for the next breakdown!
• #33: (c) 2007 Jerry Dover, P.Eng. You have all of the tools needed to quantify the benefits of implementing a maintainability program. One of the most valuable things you can do right now is ensure that repair information is captured on work orders. What is done, when, where, for how long, and what was repaired/replaced. This is your basic information for calculating MTTR.
• #34: (c) 2007 Jerry Dover, P.Eng. Looking only at labour costs, conservatively the idle labour during the downtime was costing $1,000/hr. The parts we needed were minimal, in fact some of them were surplus parts lying around. Labour was interesting – I assigned the project to one mechanic who did very little, and then gave the challenge to one of our newer mechanics, who just ran with it. I guided and suggested, but really the final design was all his. So our challenge was a 75% reduction in the time taken to repair. Note this is NOT a reliability improvement – we would expect the same number of failures each year.
• #35: (c) 2007 Jerry Dover, P.Eng. What the business should expect for supporting this project. These are the projects to pick up first: the low hanging fruit, hight-return and high-visibility projects. Sell your plan!
• #36: (c) 2007 Jerry Dover, P.Eng. The success resulted in: support for more projects, maintenance department confidence, demonstration of good practices Every maintenance dollar saved goes directly to the bottom line Many “soft savings” are harder to quantify yet still exist: Reduced safety risk Supervisory labor shifts to more value added work More focused and effective wrench time
• #37: (c) 2007 Jerry Dover, P.Eng.
• #38: (c) 2007 Jerry Dover, P.Eng. For a very detailed checklist, see the NIOSH Mining Safety and Health Topic on Maintainability http://www.cdc.gov/niosh/mining/topics/machinesafety/equipmentdsgn/maintainability/mantainability.htm
• #39: (c) 2007 Jerry Dover, P.Eng. Get Fired Up! - don’t accept the status quo Get Better! – quantify the benefits of maintainability Don’t Blink! - accept the challenge of continuous improvement