When a motor burns up on your packaging line, bringing production to a screeching halt, do you feel like it’s just another day? If you do, you’re not alone.
Food industry maintenance departments typically spend more than half of their time reacting to one emergency after another, yet plant availability may be less than 90%, says Amy Davidson, Emerson Process Management product marketing manager. “The only thing more costly than reactive ‘wait until it breaks’ maintenance is no maintenance at all,” she adds. According to Davidson, one industry study concluded that it costs 10 times more to maintain equipment in reactive mode versus predictive mode, and the cost can balloon to at least 40 times more when a failure leads to unscheduled downtime.
Davidson says the same study shows that preventive maintenance (PM) performed at given time intervals can be five times more expensive than predictive maintenance (PdM), which uses technology to monitor and recognize failure conditions before they occur.
The reason for the “wait until it breaks” mentality is lack of timely and reliable information-including the cost to the business, says Parsec Automation CEO Eddy Azad. “If accurate benchmarks were available to clearly demonstrate the impact of regular PM intervals (optimized for best performance), then it would be much easier to justify proactive PM based on facts with clear financial details,” he adds. “If you knew that replacing a cutting tool after 250 hours of operation would result in 5% less waste and rework compared to continued operation until failure, wouldn’t you do it, especially if you knew what it was costing?”
ABB’s PM30-hosted maintenance management service was used by a large bottling plant in Opelousas, LA, and produced an 11% increase in line efficiencies with 30-year-old equipment. Planned work improvement increased more than three-fold. Source: ABB Reliability Services
How processors value maintenance
Not all processors look at maintenance the same way, says Harrell Waldrop, AssetPoint executive vice president. Typically, the food industry is a mixed bag of high-value process operations and low-value discrete operations. The food industry culture has been “break/fix” because the production operation was put in place with little foresight into how to maintain the facility.Food and beverage processors typically operate on thin margins, but this should not preclude them from having a proper PM or PdM program, says Pablo Conrad, Americas industrial marketing manager, ExxonMobil Lubricants and Specialties. With a program in place, they can avoid unplanned downtime from an expensive machine. For example, a proper lubrication plan can help protect equipment and bring more productivity for the asset, and Conrad says successful processors understand the importance of following best practices in maintenance.
In companies where there is no continuous improvement (CI) initiative, the tendency is to deny downtime problems exist, states John Nichols, managing director, APEX Manufacturing Solutions. Nichols says he often hears processors say, “We don’t have any downtime in our plant.” Every plant has downtime, and it is costing them money, he adds.
For many operations it’s a cultural issue, says Ted MacDonald, Invensys Operations Management asset performance manager. Traditionally, production and maintenance have seen themselves as departments with different goals, often leading to an adversarial environment. Management needs to shift the focus to an asset performance management mindset that combines reliability and utilization, making it everyone’s responsibility.
The key issue to understand is the asset or equipment creates the demand for maintenance, says Keith Carman, AssetPoint senior account executive. “If we deal with this fundamental issue, it is much easier to develop successful plans for PM and planned repairs. Both operations and maintenance share the responsibility for non-production oriented work,” adds Carman. In a multi-site situation, each plant should normalize on a baseline built for its unique processes and equipment.
Maintenance has to be owned by the individual plants, however, it is important to have expectations driven and supported by corporate, says Dave Barberree, Rockwell sales manager. Corporate should make maintenance-related issues a key metric in gauging the success of an operation-just as important as the role that quality and safety has played in the last few years.
Besides communicating maintenance strategies and key performance metrics to the plants, a processor’s corporate group should provide adequate infrastructure (IT, systems, training, etc.), says Terry Fisher of SKF’s Service Division. Communications from corporate must include the plant manager. A key function of corporate is to ensure that the plant manager-who often has little maintenance experience-understands the corporate maintenance strategy and its impact on plant performance.
Today’s shop floor software includes maintenance capabilities and will show faults (such as stoppages and duration) reported by sensing equipment on the line, providing operators with OEE and productivity information. Source: CDC Software.
Training needed
Training is necessary to minimize downtime, and several methods can be used to provide it. According to Charlie Rogers, JBT FoodTech director of customer service, corporate should provide support with training programs on specific equipment, processes and maintenance management.A key cultural obstacle in launching training programs is the “we have always done it this way” mentality, says Fisher. “Particularly in organizations with an experienced workforce, it is critical the training demonstrates the ‘real life’ improvements possible by changing maintenance practices,” he adds. Of course, the issue of departments not working together-operations doesn’t give maintenance sufficient time to make repairs-often exists. “Although it is unrealistic to expect complete cross-training, this conflict can be improved by having a representative from operations attend maintenance training events to better understand the benefits of proactive maintenance,” says Fisher.
Rudy Westervelt, president of Power in Learning, applied a real-world, hands-on technique in cross training. “In one plant I managed, I felt the maintenance group was not taking the operators seriously. I required each maintenance technician to run a filling machine for a day with the production operator standing next to him. This turned out to be very effective, and the maintenance technicians had a better understanding of the process and the pain that occurred from repetitive problems,” says Westervelt.
For some processors, cross training occurs naturally. According to Waldrop, it’s not uncommon for lead maintenance techs to have an operator background. The maintenance tech could have an operating background, knows how the equipment fails, and can diagnose a problem without help. Waldrop points out that operator-performed maintenance is the norm for some processes and routine tasks, and it is expected where time allows and special knowledge is not needed. It’s a cost saver and a component of total productive maintenance (TPM), a piece of lean manufacturing that suggests, in part, that operators can do some level of maintenance on production equipment.
Unilever Bestfoods (Baie d’Urfé, Quebec) began a TPM initiative in 2002 that aims to eliminate all losses, striving for zero accidents, defects and failures to improve the overall effectiveness of its production system. The processor recently integrated an effective lubrication management program within its TPM initiative. While the plant already had a computerized maintenance management system (CMMS) in place, it first wanted to evaluate whether the CMMS could support adequate and user-friendly lubrication management. Second, Unilever conducted a lubrication contamination control point (LCCP) evaluation to assess the risk of contamination by the lubricants and/or lubrication practices. Finally, with the help of Shell Canada, Unilever conducted a complete plant lubrication survey to determine which lubricant to use, when and how much to use and the right time and method of application.
According to Johanne Archambault, continuous improvement manager at Unilever Canada, the lubrication survey helped the company develop an enhanced lubrication action plan. “Although we had a good maintenance program in place, Shell Canada’s recommendations helped us make improvements that will accelerate our journey to become a world-class operation,” he says.
Take the next step
MacDonald’s asset performance view of maintenance suggests three ways to start an effective maintenance program that moves away from break/fix and gets everyone involved:Set up asset performance teams made up of maintenance, production and engineering that are charged with analyzing and improving performance in their areas.
Have both maintenance and production routinely inspect equipment to identify problems early and provide basic asset care such as lubrication and adjustment.
Shift the focus from measures like downtime and production volume to more holistic measures like overall equipment effectiveness (OEE).
Westervelt suggests that the first step out of break/fix is the move to PM, which should at least include developing work orders by piece of equipment. A PM work order tracking system is important and should be tied to an effective spare parts inventory. Completion of work orders should be tracked. After establishment of an effective PM program, processors will be able to move into PdM.
Sean Robinson, GE Fanuc Intelligent Platforms global industry manager for food and beverage, reports that many of GE’s customers have moved from a pure reactive break/fix mode to a preventive regimen that involves operators, supervisors and maintenance staff-each of whom has a role to play in ongoing maintenance and “keep ready” work. These processors enjoy reliability metrics that exceed the industry average by 10 to 15%. But, he warns, a move from break/fix isn’t necessarily easy and should happen incrementally.
“Generally, the real barrier is cultural,” says Robinson. If a processor and its employees are not used to being driven by data, they can’t expect to easily leap into a reliability-based approach. The move would require an instant acceptance of a lot of automated information, such as SPC tools, alarming and event management systems, and so on.
Robinson says moving to a structured downtime/OEE solution is a good first step, as it begins the process of trusting a system to hold useful, actionable data. Comparing the failure patterns in the OEE solution with historical maintenance records usually will help processors see the value in a usage-based maintenance schedule, which is the big step in getting to a preventative regimen. From there, the additional analytics and work practice adjustments needed to move to predictive and then reliability-based approaches are no longer alien.
OEE is not just a solution for operations. While processors’ operations teams will turn to systems like OEE, which can provide visibility into where the bottlenecks and areas of downtime exist in the factory, this information also provides maintenance organizations the ability to proactively target areas that historically have caused downtime in the plants, says Matt Ruth, Avanceon director of food and beverage.
While all these tools are available, maintenance departments still need to show that they have a viable plan for keeping this higher state of readiness, says Rogers. Otherwise, any gains will be short-lived. He also suggests that processors contact their OEMs to assess the current state of their production lines and have them recommend a maintenance schedule to include in their plans. OEMs can also provide periodic inspections of equipment, evaluate maintenance practices, and refine maintenance activities.
Ruth suggests that outside services can be very useful not only in consulting, but also in providing maintenance service that includes guaranteed response times.
A large bottling plant in Opelousas, LA, contracted with ABB Reliability Services NA to improve its production output and reliability culture. After working with ABB and its PM30-hosted maintenance management service through a six-step process of collecting data, doing a physical equipment inventory, creating a PM strategy, doing a user review, balancing workloads and training the workforce, the bottling plant saw noticeable improvements in line efficiencies. After implementing the PM30 software system, the plant achieved an 11% increase in line efficiencies, and production output increased while maintenance costs remained constant.
Knowledge is power
According to Vlad Bacalu, Advanced Technology Services product manager, if a processor has good PM practices in place, the next step is to review and optimize the PM plans by introducing PdM activities, reducing intrusive tasks. The proper application of these tools can be determined through a reliability analysis and equipment repair history. An indication of a solid, optimized PM program is that at least 20% of the work orders are generated as a result of PMs.
Once processors have focused on basic PMs-cleanliness, oil level checks, pressure gage monitoring, etc.-they can begin to use the tools to move to PdM, says Fisher. These include monitoring equipment through vibration, temperature, viscosity, and more. Focusing early on the critical equipment provides visibility through the organization. It’s important to have key performance indicators (KPIs), such as mean time between failures, in place and trend the data collected from the PdM activities.
Young’s Seafood, located in Grimsby, UK, contracted with SKF for its asset optimization service to improve efficiency, speed of production and delivery in two major manufacturing units. Kevin Hamilton, Young’s engineering manager, says the process began with a client needs analysis. According to Hamilton, “The final report identified the strengths and potential weaknesses of our existing procedures. Over 40 KPIs, including monitoring, downtime, reactivity, operational risk assessment and failure analysis were used, and each was then benchmarked against SKF’s database of food and beverage sector statistics to show possible problems and areas of excellence.”
A PdM environment needs field information, which can be obtained by smart sensing devices generating information about their condition and that of the process equipment to which they’re connected. An intelligent network then relays this data to an enterprise asset management (EAM) program, which digests the data. Then it can be used for predicting equipment life and scheduling repairs, says Davidson.
EAM, CMMS, PdM and condition-based monitoring are all proven technologies that provide real savings and efficiency improvements, but only if they’re implemented by knowledgeable and experienced resources-whether internal or external, says Nichols.
With a good PM and PdM strategy in place, processors can then focus on improving the reliability of critical equipment, says Bacalu. The key to a good reliability strategy is to understand that it is a living program that needs to be continuously reviewed and improved.
Once processors have focused on basic PMs-cleanliness, oil level checks, pressure gage monitoring, etc.-they can begin to use the tools to move to PdM, says Fisher. These include monitoring equipment through vibration, temperature, viscosity, and more. Focusing early on the critical equipment provides visibility through the organization. It’s important to have key performance indicators (KPIs), such as mean time between failures, in place and trend the data collected from the PdM activities.
Young’s Seafood, located in Grimsby, UK, contracted with SKF for its asset optimization service to improve efficiency, speed of production and delivery in two major manufacturing units. Kevin Hamilton, Young’s engineering manager, says the process began with a client needs analysis. According to Hamilton, “The final report identified the strengths and potential weaknesses of our existing procedures. Over 40 KPIs, including monitoring, downtime, reactivity, operational risk assessment and failure analysis were used, and each was then benchmarked against SKF’s database of food and beverage sector statistics to show possible problems and areas of excellence.”
A PdM environment needs field information, which can be obtained by smart sensing devices generating information about their condition and that of the process equipment to which they’re connected. An intelligent network then relays this data to an enterprise asset management (EAM) program, which digests the data. Then it can be used for predicting equipment life and scheduling repairs, says Davidson.
EAM, CMMS, PdM and condition-based monitoring are all proven technologies that provide real savings and efficiency improvements, but only if they’re implemented by knowledgeable and experienced resources-whether internal or external, says Nichols.
With a good PM and PdM strategy in place, processors can then focus on improving the reliability of critical equipment, says Bacalu. The key to a good reliability strategy is to understand that it is a living program that needs to be continuously reviewed and improved.
