According to the ARC Advisory Group, 35 percent of indirect cost in a food plant is due to maintenance. All parts of the plant must be up and running if a product is to be delivered efficiently and on time. Let's face it, if your label machine is down, your efficiency is down no matter how quickly you're still manufacturing product. It should come as no surprise, then, that operational excellence and improvements to the bottom line can be impacted by improvements to maintenance function. Although each maintenance type has its strengths and weaknesses, knowing which approach is right for your plant can be the key to improved efficiency.
Breakdown maintenanceThere was a time when this "fix-it-only-when-it-breaks" mentality was the basis of most maintenance departments. The objective was to be ready to fix the problem when it was identified. In the days of low utilization, less demand for quality and more flexible customers, this was not a problem. But at the crux of breakdown maintenance is one glaring reality: Every fix is an emergency situation. And emergency equates to expense.
"Almost 90 percent of organizations believe that preventive maintenance increases productivity and the return on assets. Yet, only a third of those companies currently spends more than half of their maintenance budget on preventive strategies," says Brian Dunks, global product manager for Intentia's Movex Maintenance application. "A policy of ‘fix-it-when-it's-broken' is an expensive luxury."
Preventive maintenanceWe learn from our mistakes-or at least we hope to. And so, the industry moved from breakdown to preventive maintenance in hopes of reducing costs and improving efficiencies. The objective of preventive maintenance is to stop breakdowns and avoid unplanned downtime. Preventive maintenance programs are based on the theory that the majority of failures are a function of time.
The Unilever Best Foods (UBF) plant located in Wichita, KS, recently installed a DataStream's MP2 maintenance system to help it take control of its inventory and maintenance schedule. UBF uses a preventive maintenance system to produce about 80 percent of its work orders, with the remaining 20 percent of the work comprised of work requests and after the fact work orders (primarily emergency repairs).
Keep in mind that many studies indicate that as many as 80 percent of failures aren't based on time. In fact, in some cases, the result of time or meter-based inspections or overhauls was either too much maintenance too soon or too little too late. And, on occasion, preventive maintenance has been known to actually introduce failure.
Predictive maintenanceThe objective of predictive maintenance is to detect early warnings of a failure in progress and allow you to be proactive, applying planning, scheduling and execution in an effective manner. Maintenance triggers are typically time-based with more advanced approaches allowing production information to serve as a trigger (i.e. calibrate fill head every 100 hours of run time or 75,000 pounds of product). Predictive maintenance has proven that it can lower unplanned downtime (for approximately 20% of the failure modes).
One approach, currently used by a national beverage plant, is to use a variety of predictive technologies coupled with visual or other sensory-based manual inspections with handheld devices to determine the health of bottling and packaging lines. Using an asset reliability system called Ivara EXP, the maintenance group moved to a proactive, condition-based approach to maintenance. By consolidating and analyzing key indicators from predictive technology reports as well as other sources of information, the company can determine the health of its equipment. The system facilitated the proactive approach, enabling maintenance to conduct the right work when it was needed (rather than based on time).
Reliability centered maintenanceConsidered by some observers to be one of the great maintenance developments, reliability centered maintenance (RCM) is based on failure analysis, delving into the function of the equipment and then determining how equipment fails, why each failure type occurs, and the symptoms which indicate potential failure. RCM focuses on doing the right work at the right time. Doing work too soon is wasteful. Numerous studies indicate that as much of 50% of all current maintenance work is not required. With limited maintenance resources in most plants, this means that maintenance departments could use the time to accomplish more critical work.
A key concept of RCM is the definition of failure. With previous approaches, the definition meant that it did not work. With RCM, the definition is relative to the business objectives of the organization. Failure is defined as not meeting the performance required for the particular function. For example, if the line is producing off spec product, that is a failure condition because quality is part of the business objectives.
Reliability Practitioner and VP of Marketing at reliability solution vendor Ivara, Brian Maguire claims that asset reliability improvement projects typically result in improvements equal to 20 percent of the enterprise's maintenance budget going to the bottom line. This comes from a combination of improved profitability (from increased product availability) and reduced cost. According to Maguire, a national shelf stable bakery products company realized a substantial financial impact by implementing an asset reliability improvement project on one of its lines.
Condition monitoringPredictive and reliability centered maintenance use condition monitoring tools. Abnormal conditions indicate something is wrong with the equipment and, typically, that it will fail in the near future. Common conditioning monitoring approaches include vibration analysis, thermal imaging, oil analysis and lasers to monitor the alignment of high-speed rotating shafts.
Kim Custeau, product marketing manager for Avantis (part of Invensys), says a major salty snacks company is implementing a maintenance system with the first phase including a link between a Wonderware MES system and the maintenance system. The objective is to minimize both unplanned downtime and maintenance cost through condition monitoring.
The trend among OEMs is to include condition monitoring options in the equipment they offer. Sophisticated control panels on compressors were considered pricey add-ons in years past, recalls William "Joe" Buck, corporate vice president of engineering at Engineered Refrigeration Systems, Mobile, AL. "Now everybody agrees they can't operate without them," he says. "The latest innovation is on-board vibration monitors. Those detectors cost about $9,000 per compressor. That's not a trivial investment. Nonetheless in today's economy, that's not going to slow down adoption." For example, Bosch Rexroth provides a range products for drive, control and motion technologies. Rexroth's new intelligent drive system, the IndraDrive, includes condition monitoring and predictive maintenance intelligent firmware functionality. The supplier claims that its "early warning system" allows service and maintenance measures to be taken in advance to prevent machine breakdowns. The IndraDrive not only monitors itself, it can also independently detect any weak points within the entire system. If a weak point is detected, a warning or fault is transferred to the master control. This allows for early detection of mechanical problems and the prevention of faults and errors, resulting in reduced downtime and increased productivity.
Which style is best for you?It takes money to monitor the health of equipment, which brings us to the dilemma of selecting a cost controlled or results controlled maintenance philosophy. Monitoring means inspections, sensors, and enormous amounts of information that must be gathered and analyzed. How can we determine the right maintenance strategy for a specific asset?
To meet the objectives of lean manufacturing, evaluate the cost of failure in terms of both not meeting business objectives and any extra cost due to unplanned or even emergency repairs. By understanding the cost of maintenance for each asset, you can compare these to develop alternative maintenance strategies.
Namely, if both the cost of failure and the cost of maintenance are low, you can justifiably use the strategy of simply fixing the asset when it breaks down. If the cost of failure is low but the cost of maintenance is high, you need to minimize the amount of maintenance required. This might best be accomplished by waiting for the asset to break before expending any maintenance cost. Time-based preventive maintenance will work here as well.
Where the cost of failure is high and the cost of maintenance is low, you'll need to be more proactive and accept more maintenance cost in order to insure that the asset will not fail. Here, a predictive maintenance strategy should work well. Reliability centered maintenance should also be considered since the consequence of failure is high. Finally, if both the cost/consequence of failure and the cost of maintenance are high, you'll need to use reliability centered maintenance to optimally manage the health of the asset and avoid failures.
Ultimately, the maintenance style you choose should be in response to the storms you may face. Walk the plant floor and evaluate your assets one at a time. Keep in mind that one maintenance philosophy might not be appropriate for all assets. And above all, learn from equipment failures that do occur. You don't want lightning to strike the same spot twice.