MILL MAINTENANCE

In pursuit of "flavor-of-the-month" strategies/philosophies, some pulp and paper mills have drifted away from time-proven essentials of maintenance

BY KEN L. PATRICK, Senior Editorial Consultant

Renewed Focus on Maintenance Basics Can Put Mill Programs Back on Track

In principle, the practice of maintenance has not changed that much since the early years of the industrial revolution. From the late 1800s to the threshold of the 21st century, it has remained a top manufacturing cost item-especially in the pulp and paper industry-regardless of attempts in recent years to modernize the function through computerization, restructuring, and reorganization.

While certain modernization programs might have improved maintenance productivity in mixed degrees, at the same time, mills in search of a new focus or organizational redesign seem to have drifted away from some of the time-proven essentials or basics of maintenance. Many maintenance specialists in the industry believe that the key to getting maintenance "back on track" may require more attention to some of these simple basics, using a logical, well-structured maintenance program tailored to the needs of individual mills.

John Yolton, senior industry consultant with Indus International in Marietta, Ga., has been involved with maintenance in the pulp and paper industry for many years. He points out that maintenance has "gone off on various tangents" in recent years, and that "some people have become confused about what the basics of maintenance really are."

"There is a lot of fire fighting going on out there," he insists. "The main emphasis is on getting equipment back up and running as soon as possible, when, really, some of that equipment-with proper preventive maintenance-probably shouldn't have gone down in the first place."

In Yolton's view, a methodical maintenance "process" has several defined steps. The process, as he sees it, begins with identification and proceeds through investigation, planning, assignment, execution, sign-off, and archiving. Sometimes mills get involved in the "frills" of new maintenance philosophies or strategies and forget to follow these basic steps, "as simple as they might be," he says.

Christer Idhammar, president of IDCON Inc., a specialized maintenance consulting group in Raleigh, N.C., shares a similar view. He believes that a methodical, basic maintenance program should begin with prioritized identification followed by detailed inspection or "condition monitoring," planning, scheduling, recording, and technical/economical analysis.

Both agree that the increasing lack of attention to certain maintenance basics may be due, at least in part, to the fact that, today, nobody really wants to do "menial" tasks such as cleaning and lubrication. "Also, mills sometimes get caught up in activities such as reliability centered maintenance (RCM) and completely overlook lubrication, for example," Yolton adds.

Charles Latino, president of Reliability Center, Inc., in Hopewell, Va., a company that deals with most of the continuous process industries, believes that the focus in recent years has been on trying to catch failures early so that mills and plants can reduce their overall costs of maintenance. "I frankly believe this is fundamentally wrong as a focus," he states.

"I'm not saying that this shouldn't be done," he continues, "I'm only saying it's wrong as a focus. The focus should actually be on eliminating failures altogether. Also, industries today are spending enormous amounts of money on practices such as RCM. These approaches can be done a lot less expensively than the way they are currently being done-or at least the way that some gurus say it should be done."

MAINTENANCE BASICS. Idhammar sees among the "absolute" basics of maintenance, activities that, in effect, actually "prevent" maintenance-such as lubrication, cleaning, and certain operations practices that have to be done with some understanding of the why's and what's. "But nobody really wants to do these demeaning activities, especially cleaning, he says.

"And by cleaning, I don't mean housekeeping with a water hose. I'm referring to detailed cleaning of various components, such as electric motors, gears, hydraulics, etc. This kind of detailed cleaning, which most people will agree needs to be done but no one wants to do, will invariably lead to inspection and identification of maintenance problems. You can't clean in detail without doing inspection. This is so basic that it's almost ridiculous."

Although most people in a mill-the operators and maintenance technicians-realize that certain maintenance basics need to be done, they might not really understand why and how, Idhammar adds. This situation can usually be improved through training.

As an example, Idhammar says that most mill personnel are aware of the problems caused by improperly starting up a process line. "They know that you should avoid putting on steam too fast, because heating up a system too quickly can cause water hammer, and can result in cracked bearings, races, etc. But you would be amazed at how many people cannot explain how or why this happens. Just a little teaching and understanding can make a big difference," he emphasizes.

Ongoing attention to lubrication can pay especially high dividends, Idhammar points out, adding that "lubrication is the heart of a preventive maintenance program. If done properly, it can actually prevent the need for maintenance." By implementing advanced filtration measures to keep the oil clean and prevent moisture contamination, some mills, he says, have extended oil change periods by several years and at the same time have dramatically reduced equipment failures and rebuilds.

One such mill that did implement a back-to-the-basics maintenance program, including special emphasis on lubrication, was the Buckeye Florida mill in Perry, Fla. Ron Kinsey, corporate reliability manager at Buckeye (mill maintenance manager from 1986 until his recent promotion), explains that his mill's approach to maintenance is based on a three-prong strategy.

"Several years ago, in its conceptual stage, I drew a 'teeter-totter' model of our maintenance program. On the left is preventive maintenance, in the center is preventive maintenance, and on the right are corrective maintenance activities," he explains. The key is to keep the "teeter-totter" carefully balanced so that no excessive activities build up in any of the three program areas.

"We believe that preventive maintenance is the most cost-effective action we can take. If we can prevent failure, the mill doesn't suffer equipment downtime, repairs, lower quality, etc., and all of the associated expenses," Kinsey emphasizes. He says that this is done through detailed attention to lubrication, good alignment, careful balancing of rotational parts, etc.-"whatever it takes to make a system run well."

In the lubrication area, Buckeye has permanent filters in place on much of its larger equipment, and uses portable filters with quick-connect couplings to filter the oil on small gear drives and various other smaller pieces of equipment. Kinsey says that the use of these filters has reduced maintenance activities significantly.

For example, some worm gear drives used on the mill's brownstock washers were previously being overhauled annually. After implementation of the filter system, "we haven't had to touch many of these drives in six years," Kinsey reports. "Simply keeping the oil clean and cool has made a big difference."

Buckeye also installed desiccant bead breathers to cut down on moisture intake into the oil systems of its big gear reducers. "As these units heat up and cool down, they can pull a lot of moisture-rich air into the system. "This moisture can then condense and become a major contaminant in the oil," he explains.

The mill's predictive maintenance program includes infrared, oil, and vibration analysis techniques, along with "as many predictive tools as we can find that will help us do a better job of condition monitoring on a real-time basis. We don't want any surprises," Kinsey explains. The mill monitors about 15,000 bearing points per month.

Oil analysis is an especially important predictive maintenance tool at Buckeye. "We look very carefully at the results of oil analyses to make sure that viscosity and additive packages are right and that there are no excessive metals fines or other contaminants in the oil, and that the oil is right for the machine it is being used on," Kinsey adds.

In the corrective maintenance arena, the Perry mill uses a "clean shop"-a special rebuild area in its main shop that is air conditioned and has a "closed" environment to keep dirt and dust out of bearings, gears, and housings during rebuilds. As a routine practice, all bearings, shafts, and housings are carefully "mic'd" (use of a micrometer to determine out-of-roundness, warp, etc). All rotating assemblies are balanced and carefully put back together to ensure that the right bearing fits are utilized for each specific piece of equipment.

Close attention to these details is critical, according to Kinsey, who will be starting his own maintenance consulting operation (the Kinsey Group) out of Perry this spring. "For example, when shrinking a bearing on a shaft, there is typically only two-thousandths of an inch clearance between the rolling element and the race. If the shaft is off by a thousandth of an inch and the bearing housing is off by an additional thousandth of an inch, you're in danger of ruining the unit on startup. We believe that paying attention to the smallest of details is absolutely essential, and that, in every way, it is the key to maintenance success," he insists.

Kinsey adds: "We believe that to have maintenance excellence, you must have an excellent, well-balanced, preventive, predictive, and corrective maintenance program. If you can't keep something from tearing up, you'll have to fix it. And if you don't fix it right the first time, you'll have to repair it again-over and over. We have found that, basically, excellence means handling every detail every time."

In 1995, Buckeye Florida was awarded the North American Maintenance Excellence Award for its program at Perry. The history of this award is detailed in the sidebox item on page 48.

Latino strongly asserts that there are at least two things that industry should be doing in regard to maintenance. First, he insists, there "needs to be precision in everything you do. That means you do it once, and you do it right the first time. And that concept doesn't generally exist in industry today.

"Second, when there is failure, it should be investigated down to its very roots. And these roots are not only on the mechanical and human sides, but the management system side as well. I don't think that's currently being done in industry," he says.

IDENTIFICATION. In setting up a logical, organized approach to maintenance, identification of existing and potential maintenance problems is fundamental, Yolton stresses, adding that it is also easy to overlook or bypass this step. He makes an analogy with the automobile to illustrate key points about identification.

"As owner/operator of an automobile, you're expected to perform a certain amount of maintenance throughout the life of your vehicle-change the oil and filter, keep air in the tires, clean it, etc.," he explains. "You can make any decision about whether you will do that or not, and you generally know what the consequences are.

"The manufacturer also tells you that, periodically, you will have to take the vehicle out of service and have a skilled technician do various things to it-replace the timing belt, adjust fuel injection or valves, etc. During that outage, it is also an opportune time to tell the technician about other problems, such as squeaks, noises, failure to start properly, corrosion or paint flaking, etc.-things that you have identified in operating, cleaning, and inspecting the car.

"The same sort of things happen in a pulp and paper mill. But, in many cases, we aren't really handling the responsibilities of the owner/operator. Too often, we expect the skilled technician to identify all the problems, and do all the simple and basic maintenance activities such as cleaning, lubricating, listening, touching, feeling, etc. But identification has to be a mutual effort involving the owner/operator (the machine operators and various levels of management as well) and the maintenance technician (trained mechanics and service managers in the mill)," Yolton emphasizes.

Idhammar adds that just identifying work that needs to be done might not be sufficient. "There must be some disciplined priorities-and not emotional priorities," he insists. "Prioritizing these needs gets people to thinking more about what really needs to be done, and leads naturally to the next stage, which is investigation or inspection, or what we call condition monitoring."

In the identification stage, Idhammar explains, "it all boils down to the front line management-the front line supervisors/coordinators, team leaders, planners (if the mill has this position), etc. These people are between the upper management ranks who want certain things done and the craftspeople who do the work. If there is a breakdown in communications at the frontline management level, not many needed changes will occur."

INVESTIGATION/INSPECTION. Once maintenance problems have been properly identified, certain skilled professionals need to conduct a detailed investigation of these needs. From Yolton's perspective, this is most likely a maintenance planner, a maintenance engineer, maintenance supervisor, etc., who has troubleshooting or investigative skills and can put together a tentative plan to do something "against that identified work-whether it's a preventive maintenance action or a corrective plan."

The goal at this stage, according to Yolton, is to "shape" the identified work in preparation for the planning stage. Those people identifying maintenance problems don't necessarily have the skills or knowledge to know exactly what needs to be done in relation to those problems. "It's like passing information about automobile squeaks to a skilled technician-'it only happens when I turn to the right, etc.''' The technician can then investigate and decide more clearly what needs to be done, Yolton explains. The result is a plan of possible preventive or corrective actions and an estimate of related costs, etc.

Idhammar refers to the inspection stage as condition monitoring because it typically involves more than standard predictive maintenance analysis using infrared, vibration analysis, etc. Condition monitoring includes an array of inspection techniques, including what Idcon and others call "smart methods," i.e., simplified but highly reliable ways of checking the condition of certain equipment, "that operations people can also get involved in."

Examples of smart methods, Idhammar points out, would include checking the condition of a chain drive by just measuring between test links, or checking the position of a carbon ring in a rotary steam joint to determine certain conditions of the unit. "But whoever is doing these inspections, they need to be trained in the techniques. Typically, maintenance people are very good at preparing but not very good at even thinking about inspections. They tend to say, "Why bother if it's still running?' Everybody is always too busy with other things," he says.

But all of these things, Idhammar adds, "are done for one purpose-to discover a failure before it develops into a breakdown. And if you don't have that early identification, priority establishment, and inspection, you're never going to be able to plan and schedule effectively. And if you don't plan and schedule according to real, true needs, everything becomes urgent and chaotic."

PLANNING/SCHEDULING/ASSIGNMENT. A "golden rule," according to Idhammar, is that "you always plan before you schedule." Also, he points out that there always should be a time when a schedule is "frozen," i.e., that no further changes can be made to it. For example, with a monthly shutdown, the schedule "freeze" should be a week prior to that. If a cold outage is coming up (7 to 8 days, for example), the schedule should be frozen about a month ahead. For daily and weekly work, about 20 hours before that should be sufficient.

"I'm not so sure that a mill always needs planners, but at those that do have them, I've seen it work best when they focus only on the key, important projects (such as major shutdowns). Then, of course, the maintenance supervisor and the supervisor's team can do the planning and scheduling of the finer, daily work. In that way, everybody gets involved in planning, in one way or another," Idhammar explains.

"We believe that if you plan and schedule as demanded, you give the crafts people a better chance to do a really good job," he insists.

Yolton points out that effective planning will eliminate nonproductive actions against maintenance work that needs to be done. It also leads to the development of logical steps to performing the work, including prior purchasing of materials that will be needed to complete the job.

Some work will be of the type that it makes sense to do planning for, Yolton says. "But, unfortunately, in the real world, there are other jobs that really don't require any planning-the 'rush' jobs in particular, where a planner really can't contribute anything." So these should be "tossed out" and efforts focused only on those jobs that will benefit (productivity-wise) from planning.

A mill might not have a maintenance planner position, but the function is "still there," Yolton explains, adding that a maintenance supervisor will usually have the responsibility in that case. If a mill does have a planner position, then "obviously that person would do the fundamentals of planning, which involves knowing what the job requires in terms of labor requirements, materials requirements, tooling, etc.-and applying that knowledge to a plan for assigning and executing the work."

The assignment function, he continues, is usually the supervisor's responsibility and must involve some sound logic that matches proper skills to the work that is to be done, and places the work at a time that makes sense to do it-when materials are available and equipment and systems will be down and ready for the work (if necessary).

Execution usually will be by a tradesperson. Completion and sign-off on a job "hopefully will capture what was really done on the job and all of the costs associated with it," Yolton concludes.

ARCHIVING/ANALYSIS. Idhammar insists that it is a "total waste of time to record anything unless somebody will take the time to examine how things could be done better. This leads to what we call a technical/economical analysis, which involves such things as tracking which equipment is causing most downtime, which system or piece of machinery is incurring the highest maintenance costs (compared with estimated replacement value), etc."

And, he continues, to support these types of analyses, a mill does need a good technical database. "When we are planning, for example, a good database will allow us to find spare parts. We can use it for historical analysis, to easily see how we did it last time."

But, he warns, mills sometimes jump into buying maintenance computer systems without really considering that they have to implement them. They can spend enormous amounts of money on these systems and then find that they have "all of the store items" but not necessarily the data part of the database, which can require many years of development.

One of the most important types of information to be kept in a computerized maintenance system, according to Idhammar, is the bill of materials for systems and equipment used in the mill. "A craftsperson rebuilding a pump, for example, needs to know what bearings, seals, and impeller are inside. With a good database that includes bills of materials, the craftsperson can enter the pump number and instantly see all of the parts, and whether or not they are currently in the storeroom, and how many there are.

"Building such a database from scratch can take many, many person years. But there are ways to improve and maintain database development. For example, purchasing can always require that bills of materials be delivered with all new equipment. The mill store can make it a practice to never issue parts without tying them to the specific equipment they're being used for. These types of things can help develop and keep a database up to date.

"It's a well-known international phenomenon that planners don't plan in most cases. They're too busy looking for spare parts-in the mill storeroom, in the workshops, in the supervisor's personal stores, on the operating floor, etc. And they most often end up trying to find where they can buy parts they can't find. Each piece can sometimes take days to locate," Idhammar explains.

Kinsey reports that Buckeye developed its own in-house maintenance management system in the 1970s, but since the early 1990s has been using the Flour Daniel CMMS setup. "The system we developed in-house was fairly good, but it didn't really meet our needs. And, actually, we didn't want to be in the software writing business.

"So in the late 1980s, I formed a team and we investigated several packaged systems, finally selecting the Flour Daniel approach. I believe that if you are going to have a world-class maintenance program today, you must have a computerized maintenance management system.

"A paper system will not give you world-class capabilities, because people forget, they lose papers, and they can't manage backlogs. You must have something reliable that reminds people and can provide easy accessible historical information, and will provide meaningful analyses.

"A mill doesn't have to have a leading-edge computer system, but it does have to have a decent one to be world-class-and it has to be able to use the system's capabilities well," Kinsey says.

ASSESSMENT, OUTLOOK. In general, Idhammar believes that the pulp and paper industry has made some progress in the maintenance area during the past several years. Currently, there are many good computer systems on the market, "but there is still a long way to go in fully utilizing the potential of some of these systems. However, at least the groundwork has been laid.

"In the U.S., I see more long-term thinking now than just a few years ago, and more quality in design and construction (such as increased use of stainless steel). More maintenance people today are interested in establishing reliability groups than in the past. There is certainly an increasing interest in root cause failure analysis.

"I also see more use of predictive maintenance tools such as vibration analysis than when I first became involved with the paper industry here 12 to 14 years ago. Maintenance awareness in general is much higher today. But one of the really sad things is that the U.S. industry is still lagging behind Europe and other areas of the world in training of its craftspeople, although there are some notable exceptions.

"I've been told that as much as 50% of the U.S. paper industry's maintenance force will be retiring in the next 5 to 10 years. But most companies don't really have an effective training program in place," Idhammar concludes.

Yolton points out that the industry today is being forced to become more methodical, through OSHA requirements, ISO 9000, ISO 14000, etc. And, he adds that more of this can be expected in the future. But in spite of these initiatives, mills still tend to "firefight" the maintenance function, he says. Unfortunately, the emphasis remains on keeping the production line operating to the exclusion of almost everything else.

According to Latino, "newspapers report that 1998 isn't going to be too good a year as far as the stock market is concerned. This view is principally based on the belief that productivity in general cannot be improved much more than it currently is. I personally believe that we haven't yet touched the potential for improving productivity.

"Those who make these kinds of statements typically sit in Wall Street offices and have never been on a manufacturing plant floor. They really don't understand productivity issues. I believe that future improvements in productivity can be astonishing if industry focuses on precision in everything it does, and does a good job of investigating failure down to its roots," he asserts. n

KEN L. PATRICK, senior editorial consultant for Pulp & Paper, is president of Paper Industry Communications Inc., Atlanta, Ga.

Maintenance Scorecard

According to Christer Idhammar of IDCON Inc., as the world paper industry has progressively automated during the past quarter of a century, maintenance hours have remained basically the same.

For example, about 25 years ago, the number of worldwide paper industry maintenance employees as a percentage of total employees was around 27%. Today, that figure is about 33%. The percentage has increased due to the fact that while the number of operators in a mill has fallen, on the average the total number of maintenance workers has remained about the same, if overtime and the use of maintenance contractors are considered.

"Maintenance really hasn't changed that much in the past 25 years," Idhammar reports, although some mills do claim to have made dramatic head-count reductions. But many of these mills now incur considerable maintenance overtime and use a lot of outside maintenance contract work. The result is basically the same number of maintenance hours in a mill today as in 1973."

The maintenance worker profile has changed somewhat during this period. Today there are generally fewer mechanics in a mill maintenance organization, but correspondingly more electrical/instrumentation specialists, reflecting the increased automation at most mills.

As far as how the industry is doing in regard to professional planning and scheduling of its maintenance function, Idhammar estimates that mills are well below 50%, i.e., less than half are really doing it. This increases to about 80% for scheduled shutdown work, but the figure should be in the 90%+ range in both categories, he adds.

In regard to preventive maintenance programs, the actual execution/completion rate is around 70%, when it should be 100%, Idhammar claims. "And, most probably, of the preventive maintenance programs out there, at least 30% could be scrapped because they either are not being used or they are not providing any benefits at all."

North American Maintenance Excellence Award

Buckeye's Perry, Fla., mill won this fairly recent national award for its maintenance program in 1994, and received it a year later at the National Manufacturers Conference in Chicago.

The award actually began as a one-time event under the sponsorship of DuPont, which had commissioned the A.T. Kearney consulting group to conduct a study to find the "best of the best" in maintenance practices. Seven companies were initially identified in the mid-1980s that could meet the criteria established for this study.

After completion of the project, A.T. Kearney thought it would be good to have an ongoing recognition of maintenance excellence in various North American companies and plants. They then asked each of the seven original, best-of-the-best companies to seat one member on the board of directors for a new award program.

This board of directors established the new criteria and selected its first award recipient in 1990. Since then there have been nine winners of the award, including Buckeye Florida. Each of these nine winners has been given one seat on the board of directors, as will future winners.

Whenever the number of directors becomes too large for an effectively functioning board, members from the original seven will be dropped. But so far, this has not been necessary.

We believe that to have maintenance excellence, you must have an excellent, well-balanced, preventive, predictive, and corrective maintenance program.-Ron Kinsey, Buckeye Florida