Robert C. Baldwin, CMRP, Editor

• I think I have an answer to your question about upper management not considering the consequences of their actions. It's my observation that many companies simply are more interested in the bottom line for the next quarter than they are for the longer term. Wall St. seems to expect no more and no less.
• I've seen the problem you discuss first hand. Quick MONEY! If you are a manager and can reduce/eliminate a cost that directly goes to and improves your bottom line it also goes to your bonus. It's simple, measurable and almost immediate. Rarely do I see deferred compensation at the plant level, so there are few if any negative consequences for the manager. Do this for a few years, before the neglected assets deteriorate or fail, and you've probably moved on to a promotion. If the plant you left behind starts to fail, you look like an even bigger hero because you did what the existing management couldn't do [keep the plant running].
• It's my view that our industry's problem is the lack of executive management recognition and understanding of the benefits available from the use of advanced maintenance management technologies. This is the result of users, vendors, professional societies, the press, etc., not being successful in communicating our story to high level, resource allocation, decision-making executives.
• I believe that maintenance managers must get more involved in the financial picture of their organizations. For upper management to--get it,--maintenance managers should make clear to the decision-makers the old cause and effect rule. This requires tracking all maintenance activities as they relate to the productivity of the organization. Unless the connection between the value of maintenance and the bottom line is well established, maintenance is just another expense to minimize. This connection is made through factual, well-presented financial data. Upper management is less likely to act irrationally when well aware of the value and interdependency of a well-managed and documented maintenance organization. The successful maintenance manager of today must hold both technical and financial degrees. It is no longer good enough to know the nuts and bolts; we must now know bytes and beans.

Perhaps we are partially to blame. Time to start counting those beans. MT

Robert C. Baldwin, CMRP, Editor

I related his comment to the predictive maintenance technicians in each audience, asking: How does that make you feel? What is a guru, anyway? Do you really want to be one? Is that good or bad? For the second question, I had a ready dictionary answer:

gu·ru Hinduism. A personal spiritual teacher. 1. A teacher and guide in spiritual and philosophical matters. 2. A trusted counselor and adviser; a mentor.

Some of my remarks, I hope, contained an idea or two that would help members of the audience to answer the other questions for themselves.

I wish I would have had the foresight to discuss the issue further with the maintenance manager who made the guru comment. What did he mean by guru? And did he think a guru was good or bad? From his tone, I assumed he equated guru to an expert with an attitude.

There are a number of functions or roles in an effective condition monitoring or predictive maintenance program. Guru is just one. Others that come to mind are champion, user, analyst, and field technician.
The field technician makes the inspections and collects the data and samples to be studied by the analyst, who makes recommendations to the user, who will cause the maintenance organization to take appropriate action on the information. The champion is the person who may have introduced the concept and currently supports it with enough strength to keep it going. The participants' knowledge is derived from the teacher, expert, or guru.

These functions may be fulfilled by a team or a single person, by in-house personnel or an outside service organization, but they must be fulfilled and be part of a rational process if condition monitoring is to be successful. Some valuable guidance on how to make the process work is covered by Jack Nicholas and his associates in the article on Strengthening Your Predictive Condition Monitoring Program (page 12).
In my mind, to become a guru is a worthy goal. The true guru is wise enough to know that all the functions must be fulfilled, even if they have to be done by the guru himself. MT

Bob Williamson

When lean gets interpreted as downsizing by many of today's business leaders, they make the mistake of reducing headcount in their organizations to make them leaner from a staffing perspective. Well, that is not the intent of lean.

A fundamental characteristic of a lean organization or lean manufacturing is the systematic identification and elimination of waste to reduce manufacturing or operating costs. Targeted forms of wastes are associated with overproduction, transportation, motion, inventory, processing, defects, and waiting.

Unreliable equipment also represents a significant waste--extra inventory to compensate for breakdowns; extra backup equipment; processing delays due to unplanned downtime or inefficient performance; defective materials produced due to breakdowns; waiting for information, parts, and materials to make needed repairs; or waiting caused by inefficient equipment operation. Eliminating equipment-related wastes (or losses) is fundamental in achieving the goals of lean.

If the organization's leadership assumes that lean means fewer people and begins reducing headcount without eliminating, or at least reducing, equipment-related waste, an upward cost spiral begins. With fewer people to respond to equipment problems, or to perform required preventive maintenance, equipment performance levels and reliability suffer even more. This approach can actually increase manufacturing or operating costs rather than reduce them.
Downsizing and lean are not the same! Downsizing, without eliminating waste, is typically not sustainable. Rather, it is a one-time, short-term cost reduction strategy that if left alone will likely lead to increased costs.

So, what are the correct methods for becoming lean in a sustainable manner? Begin by identifying the types, reasons, and root causes of waste that have a direct and immediate impact on business performance. For equipment-related wastes be sure to involve the people closest to the problems--maintenance and reliability (repairs and prevention), operations/production, purchasing/stores (repair parts), and engineering/technical (design and modification). Identify and eliminate the causes of poor performance using formal problem identification and root cause analysis methods.

This takes data. Some organizations have excellent data, which makes this step easier; with very sketchy data, this step is difficult. In the absence of data go with what you know. Baseline the targeted equipment performance measures and then begin collecting data to measure equipment performance and if improvements are actually being made.

Identify action items to correct and eliminate the root causes of poor equipment performance. Look at equipment conditions and data. Look at work processes and procedures used to operate, maintain, and document changes, control quality, communicate, and schedule anything to do with the targeted equipment. Consider the people who directly, and indirectly, affect the performance of the equipment--their qualifications, training, and numbers.

What then are the essential elements of becoming lean in a manner that is sustainable?

• There must be a clear and compelling, and urgent, reason to change.
  
• Cross-functional leadership must proactively and visibly lead the organization through the change process.
  
• Leaders must continually communicate, and role model, the new vision and strategies.
  
• Leaders must break down barriers to making necessary improvements.
• Leaders must engage the people closest to the top priority problems or opportunities to identify, design, develop, plan, and implement improvements.
• Leaders must leverage the successes and key learnings for making improvements by eliminating waste in other areas.
• Leaders help everyone in the organization understand the connection between the improvement activitires and results with the vision of the organization so the new behaviors become part of the "way we engage our prople and run our business."

Those of us in maintenance and reliability roles can help organizations become lean by targeting equipment-related wastes and keeping our business and labor leaders informed of the results. MT

Robert Williamson of Strategic Work Systems, Inc., Mill Spring, NC, is an author, workplace educator, and consultant with more than 27 years' experience in improving the prople side of manufacturing and maintenance with many fortune 500 companies.

Condition monitoring (CM) and computerized maintenance management systems (CMMS) have evolved and coexisted as separate disciplines. Many efforts have been made to integrate CM and CMMS with nominal success.

Conceptually, CM and CMMS have different functions and their applications yield different results. Understanding these roles provides significant insight into the benefits of an integrated system. In general, the role of CM is to implement a maintenance strategy and the role of CMMS is to manage the execution of maintenance. These separate disciplines have been successfully practiced for years, each on its own merits with relatively little knowledge or interaction with the other. When considered as an integrated whole, it becomes clear there are exciting possibilities for greater benefits.

Benefits of integration
The integration of CM and CMMS provides clear opportunities including:

• More effective and automated implementation of maintenance strategy. Research has shown that condition-based maintenance provides the lowest maintenance cost and highest availability for many plant assets. In practice, these benefits can be elusive. Effective communication of CM recommendations and tracking the results provides a powerful tool to support complete realization of the CM benefits. Meaningful communication between CM and CMMS provides automatic, paperless execution of CM, minimizing man-hours and increasing effectiveness. This connection between the systems allows a maintenance strategy based on machinery health to be institutionalized as a part of the user's business.
• Improved accuracy of CM analysis. Communication of information between CM and CMMS improves CM analysis in two important ways. First, it allows the analyst to observe the work history of the machine being analyzed. Armed with this knowledge, the analyst can recognize the difference between a new bearing that may produce high readings as it wears in and an older bearing that will produce high readings as it degrades. Most maintenance actions will impact CM measurements, and understanding this work history results in dramatic improvements in CM analysis.

The second improvement in CM analysis is from systematic feedback on CM recommendations. The two most important outputs from CM are diagnosis and prognosis. Diagnosis identifies what is wrong with the machinery, and prognosis estimates how bad the condition is or, ideally, answers the question, How long will it last? With an integrated approach to CM and CMMS, CM recommendations are tracked and the actual findings documented. This provides a tool to confirm the diagnosis and prognosis that CM generates. Prognosis based on CM information is not an exact science and it will depend on the site-specific application of the machinery. Tracking CM recommendations and supporting them with factory floor or shop observations is a powerful tool for improvement.

• Identification of repetitive failures for root cause analysis. At many plants, the biggest savings opportunity is designing out repetitive failures. In most cases, this doesn't happen because these repetitive failures either are not noticed or are tolerated by an adaptive maintenance philosophy "Oh yeah, that machine breaks every six months." CM alone can be very effective at identifying this problem and CMMS can make fixing it efficient. When the two work together, the repetitive nature of the problem and the associated costs become apparent. The measurement tools used for CM also often can be applied to study a repetitive failure and identify its root cause for design-out consideration.
• Effective communication of machinery health throughout the enterprise. Availability of machinery health information from the CM system throughout the enterprise creates the opportunity for significant benefits in production, engineering, and other business segments. When this understanding of machinery health becomes institutionalized, production schedules can be optimized, selection and design of plant machinery improved, and maintenance practices fine tuned. Providing the tools for continuous improvement of plant operations, generally, and the maintenance function, specifically, is the biggest benefit to integrating CM with CMMS.

How these systems work together
The premise of CM is that carefully selected measurements made on a regular basis can show machine condition accurately. With this understanding of machine condition, specific maintenance actions can be carefully planned. Maintenance interval and machine availability are optimized, driving maintenance costs down and production up. The CM domain has evolved in a technical fashion wrapped around measurement technology. Measurements can range from simple parameters such as temperature, pressure, or flow to complex data such as vibration spectra or infrared images. In all of these cases, the objective is to determine what is normal for the machine, how much change is allowable, and what the changes indicate. In practice, CM has a well-developed vocabulary and data set including:

• Plant machinery hierarchy
• Criticality
• Measurement locations
• Measurement definitions
• Measurement interval
• Severity
• Alarm status or exception
• Trend
• Spectrum
• Time waveform
  
• Thermographic image
• Frequency component
  
• Diagnosis
  
• Prognosis CMMS

CMMS also has been practiced for decades. It is an information-intensive application offering significant benefits through gathering and distributing information about the maintenance function. Managing maintenance information has been a driving force in this development. CMMS also has a well-developed vocabulary and data set that includes:

• Plant machinery hierarchy
• Work requests/orders
• Work plans
  
• Work schedules
• Labor resources/costs
  
• Parts inventories/costs
  
• Storage locations
• Preventive maintenance actions
• Purchase requests/orders
  
• Safety procedures

Creating an intelligent connection
It is ironic that two disciplines such as CM and CMMS that are practiced, in many cases, by the same people fulfilling their assigned duties, have such little overlap in the data they handle. In fact, the biggest overlap is probably plant assets, represented within both systems as a machinery hierarchy. Unfortunately, these hierarchies usually develop at different times to fulfill different purposes and they have little direct connection. The challenge in achieving greater efficiencies through connecting these systems begins to emerge. Although there are visible synergies to pursue, in most cases there is no inherent commonality between the systems. Each of these tools operates in a different domain with different data of interest and vocabularies.

Failure to recognize this challenge has been one of the root causes for the limited success of many efforts to integrate CM and CMMS. In order to address this effectively, it is necessary to effectively connect the shared data between these systems and establish new methods for the systems to exchange other information that will allow users to realize the potential benefits. The approach presented here establishes these new types of information and relationships between the systems:

• Connection between the machinery or asset hierarchies of CM and CMMS
  
• Creation of a new CM result known as Advisory
  
• Creation of work requests based on Advisories
• A gateway to automate communication between the systems
  
• Tracking work requests within the CM system
• Display of equipment histories and work plans within the CM system

The role of people
The integration of the business processes needs to be driven by the organization and its business requirements, not the software. The organization should, however, take into consideration the functionality within the software and database platforms in order to achieve integration as simply and straightforwardly as possible. Although it is a common objective to minimize the human effort required, integration does not necessarily mean without human intervention.

The integration presented here recognizes the expertise of the CM analyst and CMMS maintenance planner. It provides meaningful automation of the work request process for the CM practitioner, but it in no way attempts to create work orders automatically from gathered data without human intervention. The CM systems available today provide useful diagnostic tools to assist in the recognition of machinery faults and specific defects. Armed with that information, it is a straightforward task for the analyst to confirm the diagnosis and submit the work order using the gateway to CMMS. This gateway then offers a view of this work request as it is processed and the maintenance is executed. This makes the work process visible to the CM practitioner, ensuring follow through on his recommendations and feedback on the entire maintenance process. MT

Then why is the wind often ignored when performing thermographic surveys? Most thermographers simply do not know how important wind is in cooling down a hot spot. Also, how should they compensate for convective cooling effects? This article presents some interesting data using a simple experiment of blowing air on a hot spot simulated on a fuse cutout.

The sun also can be a strong influence on outdoor thermographic surveys from both reflective and warming standpoints. Solar reflective effects have been widely discussed. Use of long-wave cameras (8-12 mm) is the optimal solution for solar reflection problems. With short-wave cameras (3-5 mm), thermographers have had good success by changing position with respect to the target, surveying at night, and learning to interpret reflections.

Solar warming can be a more subtle effect, especially for hot spots that are thermally isolated from the surfaces the IR camera sees. For these indirect targets, temperature rises of a few degrees Fahrenheit can indicate significant problems. Transient solar loading can wipe out these small temperature rises and they will not be seen.

One utility found that great care must be taken when performing thermographic surveys on underground equipment that is heavily electrically insulated. The electrical insulation also serves well as a thermal insulator, making these underground components indirect targets. Just a few minutes of exposure to sunlight made thermography impossible on these underground components. It is possible that by waiting long enough for thermal equilibrium, perhaps several hours, that the rise due to the internal problem would be re-established on top of the solar loading. But most thermographers do not have that kind of time. It is simpler and quicker just to shield the components from direct sunlight.

For indirect targets that soak in sunlight such as oil filled circuit breakers (OCBs), thermographers need to compare apples to apples--that is, be sure when comparing OCBs that they are equally solar loaded and have been for some time. More work needs to be done in this area, but thermographers have had success in documenting major problems indicated by small temperature rises for equipment in full sunlight.

Wind effects We set up an experiment in our student laboratory that allowed students to vary and measure wind speed blowing on a simulated hot spot on an actual fuse cutout. We recognized the possibility of deriving some good data from this experiment. We were able to control wind speed from 1 mph to more than 30 mph. A squirrel cage blower provided the wind onto a Type XS 14.4 kV 100 A fuse cutout. The wind was aimed at the top of the cutout, nominally centered on the knurled brass piece. We taped Scotch Brand 88 black vinyl electrical tape to this piece to increase the emissivity to 0.95 and to attach a type K thermocouple.

Regulated 18 V dc variable power supplies provided power to both the squirrel cage blower and the heat source mounted internally near the top of the fuse cutout. We used a pocket wind meter to measure wind speed. We first heated the cutout without any wind, allowing 1 hr to attain thermal stability.

We did experiments with initial temperature rises varying from 130 F down to 45 F by varying the power to the heat source. We then applied power to the squirrel cage blower to achieve various wind speeds ranging from 1 mph to 25 mph. Temperatures were measured with both an IR camera and a dual thermocouple setup.

The experiments show for several power inputs that the influence of wind is quite strong, even for low wind speeds. The temperature rise was cut in half with just a little over a 3 mph breeze. The stronger the wind, the cooler the hot spot, up to a point. As the curves show, the largest changes occur at lower wind speeds. Our data show that between 50 and 55 mph, the wind has cooled the hot spot to ambient for the power levels we used.

Cooling by convection depends on many factors, not the least of which is shape. The size, shape, orientation to the wind, and surrounding structures all affect convective cooling. Whether the hot spot is emanating from a recessed area in the component as is often the case with hinges, for example, could make a tremendous difference in interpretation. Such a region may be shielded from the wind, and largely unaffected by it.

What does all this mean for thermographers? Here are our recommendations for dealing with wind, whether from natural sources or generated within your facility:

• Buy an anemometer. Pocket size units are quite accurate and cost about $100. Use it on surveys. Note that getting the actual wind speed on the hot spot can be difficult. Do not place the anemometer within inches of energized equipment. Try to get enough measurements to ensure confidence of the range of wind speeds the hot spot sees. Recognize the shape and orientation of the hot spot component relative to any surrounding structures. These factors strongly affect wind effects.
• Within a facility, blowing air can affect measurement on components inside normally closed cabinets. Opening the door can allow cooling air to enter. We have found some hot spots can be significantly cooled this way. If there is air blowing on cabinet doors, we recommend shooting them just after opening, before cooling can take place.
• If possible, measure component temperatures on the leeward (downwind) side of the hot spot. There will be a temperature difference from the windward to the leeward side of the hot component. Measuring out of the wind gets you closer to the no-wind condition.
• "f you are using severity criteria, find out if they are for no wind or light breeze. If they are for no wind, even a slight breeze can throw you off by a factor of two on temperature rise. " The higher the DT for a given wind speed, the higher the power dissipation in the hot spot. For a 100 A current to generate 30 W of power, the resistance would be 3000 micro-ohms.This resistance level would be a problem in medium- to high-voltage circuitry.
• We did all measurements under steady-state conditions. Steady state means the heat capacity (thermal mass) of the component does not enter into the physics of what is happening. When making measurements in a variable wind, or if the wind changes from high to low or vice versa, this is nonsteady state; the heat capacity of the component must be considered. This complicates matters considerably. High heat capacity components will be slower to heat up after the wind dies down and slower to cool down when the wind picks up.

Solar effects The sun can be a great help to thermographers in transient heating/cooling applications such as roof moisture surveys. For steady-state heat flow applications, the sun can cause problems in measurement. The effect of solar reflection creating false indications or masking true hot spots has been widely discussed. In this article, we will concentrate on the effects of solar loading on indirect measurements, particularly those underground components that the sun illuminates only when the thermographer opens a door or cover.

When normally closed compartments are opened, environmental effects such as airflow mentioned above and the sun if outside may cause problems. Underground switchgear is normally heavily insulated. A hot spot simulated in a high voltage elbow, a typical component, with an internal temperature rise of 133 F as measured by thermocouple, has an external hot spot temperature rise of only 17 F.

Heating was simulated with an internal source under laboratory conditions. In this condition, a thermographer aware of indirect measurement criteria would easily determine a problem condition.

But what happens if we allow the part to be warmed by the sun? We did not calibrate the lamp to deliver exactly equivalent solar radiance to the elbow. Rather we wanted to show that the effects of solar warming as the variation in ambient solar radiance can be considerable. The lamp delivered more energy than would the sun. However, we have observed this effect under actual solar loading conditions.

To compare our results of solar warming of both a good and bad elbow, we added a good elbow to the setup. The good elbow is at an angle and slightly above the bad elbow. With the sun shining on the elbow, there is considerable glint or reflection, as a 3-5 mm bandpass camera was used. After warming, we shielded the elbows from the lamp. In both cases, we could not tell the good elbow from the bad elbow. The heating by the lamp with or without the glint masked the problem. The lamp was on only for a few minutes. Lamp intensity was greater than that of the sun, but our experience has shown it takes only a few minutes for actual solar effects to produce similar effects.

The bottom line is that for normally shaded components where the problems are indirect thus low temperature rise, the sun should not shine on them.

Indirect measurements where the hot spot is thermally isolated from the surface viewed by the camera are more susceptible to wind and sun than direct measurements. They have a much lower temperature rise and can be masked more easily. Attempting to quantify these effects can result in some degree of frustration. Even under controlled conditions there are many variables to consider. Documenting electrical load, wind, and sun conditions can go a long way to help trend problems over time. MT

This article is based on a paper presented at the Predictive Maintenance Technology National Conference, November 15-18, 1999, Atlanta, GA.

Robert Madding is manager and Bernard R. Lyon, Jr. is thermography course moderator, at the Infrared Training Center, North Billerica, MA, telephone (978) 901-8405; Internet www.infraredtraining.com; email michelle.mcdonough@flir.com

How does your income match up with others in the maintenance and reliability community? It may be hard to find out where you stand because income figures vary so widely almost any way the data are tabulated. That is what MAINTENANCE TECHNOLOGY Magazine found out in its first two surveys of reader income. This year, respondents' income ranged from $22,500 to $112,000.

Average income of all readers responding to the survey was $63,365, somewhat more than the $58,748 registered last year. Salaried readers averaged $67,354, while readers paid on an hourly basis (22 percent of the respondents, the same as last year) averaged $49,182.

The survey was conducted over a random sample of magazine readers (except for subscribers affiliated with consultants and contract services), and we believe the data are representative of maintenance and reliability leadership.

Salaried personnel often have worked in the maintenance crafts or trades, as is the case with 44 percent of this year's respondents. Overall, 63 percent of survey respondents have worked in the maintenance trades or crafts: 50 percent as electricians, 65 percent as mechanics, and 36 percent in more than one trade, including HVAC/R technician, millwright, pipe fitter, and stationary engineer.

Age and income profile
The first of the accompanying charts are histograms of age and income of survey respondents. The age chart, with frequencies displayed in 5-year increments, shows about half the respondents were between 45 and 55 years old, with the midpoint and average close to 46. The income chart shows that about half the respondents received between $50,000 and $75,000 in annual income. The midpoint was $61,300, slightly lower than the average of $63,365.

The scatter chart that plots income versus age grouped by decades shows the wide variance of income within each of the groupings. Average income rose with age from about $43,000 for respondents in their 20s and then leveled out at slightly more than $66,000 when respondents reached their 50s.

How do respondents feel about their level of compensation in relation to their job responsibilities? More than half believe their pay is about right (53 percent), and a few felt it was generous (4 percent). The rest thought their pay was too low. Compensation policies Several survey questions dealt with compensation policies. First, respondents were asked if any of their pay was based on performance and, if so, in what sectors. Personal or individual performance led the list at 51 percent, followed by company performance at 41 percent, department performance at 27 percent, and team performance at 13 percent. (In most cases, percentages have been rounded to whole numbers, and they may total more than 100 in some categories where multiple answers were possible.)

Nearly 46 percent of respondents received a bonus last year. (Total income, including bonus income, was the income figure used throughout the study.) Of those receiving bonuses, 62 percent had some amount of their income at risk and subject to reduction if certain conditions were not met.

Education and registration
As expected, average income rose with the level of education. More than 63 percent of the respondents indicated they had some type of college degree. Average income rose from $54,232 for respondents with associate degrees, to $69,267 for respondents with bachelor degrees, and to $74,063 for respondents with advanced degrees. Average income for respondents not reporting their education level was $58,526, slightly above those with associate degrees. The chart showing income by education level shows that although the trend of the average is upward with increased formal education, there is a wide spread within each grouping. Slightly more than 17 percent of respondents were registered professional engineers (17 respondents) or certified plant engineers (18 respondents). Average income of this group was higher than the average income of all respondents. Professional engineers received an average income of $73,254 while certified plant engineers received an average income of $59,862.

Involvement and responsibility
All respondents were involved in or responsible for plant equipment maintenance and reliability. That is the basic qualifying question on the application to receive MAINTENANCE TECHNOLOGY, and all respondents receive the magazine. However, they work at different levels and have varying responsibilities within the enterprise.

Respondents were asked to choose their level of involvement. Average income was $84,055 for corporate or multiplant involvement, $56,023 for plant management level, $66,013 for maintenance or reliability manager level, $65,049 for supervisor level, and $55,624 for maintenance engineer or technician level. The chart shows a wide spread of income within each involvement sector.
Respondents also were asked about their job responsibility in three broad sectors (12 separate categories):

• Managing responsibilities sector: Department performance, hiring maintenance personnel, budgeting, and time management/supervision of others.
• Designing/buying responsibilities sector: Engineering/design, management of contract services, ordering or specifying plant equipment, and ordering or specifying tools or supplies.
• Hands-on responsibilities sector: Hands-on planning of maintenance work orders, hands-on predictive maintenance analysis, hands-on troubleshooting of equipment, and hands-on maintenance or repair of equipment.

The average respondent had job responsibilities in six of the 12 categories, with 59 percent having responsibilities in all three sectors. When grouped by sectors, 81 percent of respondents had some responsibility in the managing sector, 91 percent had some responsibilities in the designing/buying sector, and 80 percent had some responsibilities in the hands-on sector. Average income for persons having some responsibility in the sectors was $65,407 in managing, $62,621 in designing/buying, and $59,269 in hands-on.

Income by industry
The industry classifications on the MAINTENANCE TECHNOLOGY Magazine qualification form were used to learn which industries were represented in the study. Results were combined into four general sectors (processing, manufacturing, utilities, and facilities) to facilitate analysis. Average income for industry sectors was $70,360 for process industries, $63,267 for manufacturing industries, $68,887 for utilities (electricity, gas, water), and $55,837 for facilities (government, hospitals, colleges, office buildings, etc.).

Apparent satisfaction People involved in or responsible for equipment maintenance and reliability tended to be satisfied with this profession and their employer. Their average tenure is 20 years with their present employer and 17 years in the reliability and maintenance field. When asked if, when looking to the future, they would recommend maintenance and reliability work as a career choice, the answer was yes by nearly a 10 to 1 margin. MT

Questionnaires were sent to a random sample of MAINTENANCE TECHNOLOGY Magazine's 53,000 readers, minus those involved in consulting and contract services. A total of 216 responses were received and processed. No monetary incentives were used (however, respondents who faxed or sent their name and address separately will be provided a copy of the results). Averages and other indicators were based on a sample size that varied because all respondents did not answer all questions.

Judging from comments received via e-mail, expressed on maintenance-oriented web sites, and repeated in Bob Baldwin's Uptime editorials in the July/August and September issues of MAINTENANCE TECHNOLOGY, there seems to be general agreement in at least one area within the maintenance and reliability community. Maintenance professionals are concerned about the stature, respect, consideration, and response received from both their own and supplier organizations. Although there appears to be a lot of complaining, there does not appear to be much real corrective activity.

The primary issue The general question is how do we as a community change what most agree are discouraging conditions. The real question is what are you doing personally to drive change?
Before the answer, some observations: For a soon to be released book on asset management, I personally reviewed over 100 papers presented over the past two years. Out of more than 70 pages of notes, I had a little more than two pages on results and benefits. Many authors seem to be talking about what to do and how to do it, but they seem to be ignoring the benefits that have been achieved. Is that because process is more important than results to maintenance professionals? If so, priorities are reversed.

If my experience is typical, comments to Viewpoint opinions are always well thought out--but are few in number. This observation parallels that of others. A good friend who writes monthly editorials for another magazine seldom receives any comment--even to controversial subjects. On one occasion he offered more detailed information on a subject of great interest. Only a few replied. One of the replies was so far removed from his offer that he had to go back to the editorial to make certain he had been clear. Do these observations mean that many members of the maintenance community like to complain but few are willing to take the next step and become personally involved driving a solution?

We seem to be a vast army, largely unorganized and content to complain among ourselves and on web sites. We complain to the choir in ways that will never attract the attention or interest of a manager or anyone positioned to do anything about the complaints.

Is there a solution? There are at least three organizations capable of driving a process to gain greater respect and stature for the profession: the Society for Maintenance & Reliability Professionals (SMRP), Association for Facilities Engineering (AFE), and the American Society of Mechanical Engineers (ASME), Plant Engineering and Maintenance Division. These three organizations probably have a total membership of 16,000 to 20,000 maintenance professionals.

That leads to more questions: why isn't membership larger and are you an active, participating member? If you are a member, are you weighing in with your successes and results and your requirements for the changes that must be made to gain recognition and respect for your contribution? If not, please don't complain about how managers and suppliers don't listen. While they may not listen to you individually, they will have to listen to 15,000 yous, especially if the message is conveyed through an influential organization known to represent a consensus of active, energized members. If you are complaining but don't belong to any of the professional societies, don't help set their agenda and drive their efforts, don't participate in any conferences, and don't publicize your requirements and successes. I suggest you need look no further than the nearest mirror to find the person who could make real change occur. Some are probably saying I don't have the time to participate, can't afford to join, my boss won't allow me to attend conferences, and my company won't allow me to publicize success. Again I'll say you must do something more than complaining individually for change to occur. Some are finding the time--more must do so.

Your participation is mandatory not just needed. With your active participation and leadership, changes are not only possible, they will occur. MT

For more information on the professional societies mentioned, visit their Internet sites:
Society for Maintenance & Reliability Professionals, www.smrp.org
Association for Facilities Engineering, www.afe.org
American Society of Mechanical Engineers, www.asme.org

Robert C. Baldwin, CMRP, Editor

According to the U.S. Department of Labor, by 2006 there will be 151 million jobs in America but there will be only 141 million workers. Although this may be good news for the technical knowledge worker, it is bad news for reliability and maintenance organizations.

Some managers are already wrestling with the problem. In fact, the Maintenance Excellence Roundtable, of which Maintenance Technology is a member, devoted part of its conference agenda to the subject. (Other members of the Roundtable are Alcoa, Allied Signal, Baxter Healthcare, Dofasco, Dupont, Exxon, Ford, Kodak, Novartis, Sunoco, and U.S. Postal Service.)

As was pointed out at the Roundtable, you will not be able to cover the shortfall by calling on contract service organizations because they draw their skilled maintenance workers from the same labor pool as your company.

Companies bidding for workers with scarce skills will fuel employee turnover. One important issue will be figuring out how to hang on to the good people you already have. Important insights to this retention challenge have been published by Kepner-Tregoe (www.kepner-tregoe.com), a Princeton, NJ, management consultancy, in its research monograph Avoiding the Brain Drain: What Companies are Doing to Lock in Their Talent.

Kepner-Tregoe identified 11 retention leaders and derived a number of drivers of retention success from in-depth interviews with these companies. Several caught my attention:

• Retention leaders don't manage retention, they manage people. And they view people management as a strategic business issue. They know that employee knowledge is the only sustainable competitive advantage.
• Retention leaders are relentless in their pursuit of continuous improvement. All of these companies keep asking questions, soliciting feedback, and taking actions to maintain a high level of satisfaction among their workers.
• Retention leaders have a culture of caring, balanced with a tradition of excellence. On one hand, they place a high value on integrity, ethical behavior, and truth in all their dealings, including their treatment of and communication with their employees. On the other hand, they all have a rock-solid tradition of holding employees to a standard of business excellence.
  

Ultrasound equipment can identify compressed air leaks so they can be repaired before they result in unscheduled downtime, affect product quality, pollute the environment, or endanger people's lives. Photograph courtesy UE Systems, Inc.

Delaying the replacement of a leaking $100 steam trap could waste $50 a week or $2,600 a year. Since an average facility has hundreds of steam traps, leaking ones may be squandering hundreds of thousands of dollars annually. In addition to wasted dollars, unattended leaks also may result in unscheduled downtime, affect product quality, pollute the environment, and endanger people's lives.

This article deals with leaks from three different perspectives: detecting and pinpointing leaks before they mushroom into major trouble, using mechanical adhesives to repair leaks, and installing hardware to prevent leakage and improve equipment reliability.

Early leak detection Ultrasonics has been industry's technology of choice to detect and pinpoint leaks for more than 25 years. Inspectors using a hand-held, battery-operated ultrasound instrument, such as the Ultraprobe 2000, from UE Systems, Inc., Elmsford, NY, can hear leaks in vacuum or pressurized systems as well as faults in operating machinery and electric transmission and distribution systems. State-of-the art accessories such as close-focus modules and stethoscope extensions enhance the capabilities of ultrasonic instruments.

An ultrasonic detector senses subtle changes in the ultrasonic signature of a component and pinpoints potential sources of failure before they can cause damage. Longer wavelengths of lower-pitched sounds are gross waves that can be difficult to locate. But higher frequency sounds are short wave signals localized to the source of emission. For this reason, it is possible to use ultrasonic sensors in relatively noisy environments.

Continuous monitoring. While most applications for ultrasonic inspection are focused on hand-held portable instruments, there has been increasing interest in continuous equipment monitoring.

Continuous monitors include two basic components: a processing unit and a sensor, which often is in the contact mode. A wave guide is affixed to a set test point by either bonding it to a surface or by drilling a threaded hole and screw-mounting the wave guide on the object to be monitored. The processing unit may feature adjustments for sensitivity/dB level, a threshold setpoint for alarm, and outputs such as 0-10 V dc, 4-20 mA. Some units provide a heterodyned signal which allows remote listening or downloading to recording devices such as vibration analyzers, tape recorders, or computers.

An example of this type of monitoring device is a valve leak onset alarm. When a valve is shut, there is no sound. A baseline is set when the instrument is installed. If the valve leaks, the onset or increase of sound intensity over the set threshold will set off an alarm. The generated sound is usually localized to the test area where the sensor is affixed. This reduces false alarms produced by irrelevant sound generation.

More than a billion gallons of industrial fluids are wasted through leakage every year. This hydraulic pump leak can be repaired using anaerobic thread sealants. Photograph courtesy Loctite.

Since the ultrasound event produced by these emissions can be detected at a distance, these detection devices provide safe scanning around potentially hazardous high voltage equipment. Low level emitting leaks are a different problem. The signal amplitude is extremely low and needs some form of amplification beyond the normal range of most standard microphones. Receptors to enhance low level leak emissions have been developed and offer a reliable method for locating these leaks.

Liquid leak amplification. When low level leaks do not produce turbulent flow, it is not possible to detect them with conventional scanning probes because ultrasonic leak detection of either pressure or vacuum leaks depends on the generation of a turbulent flow as the gas moves from high pressure to low pressure. When it is not possible to locate this type of low level leak (typically below 1x10-3 ml/sec), using a liquid with a low surface tension will help. Only a small amount of liquid must be applied to the leak test area. As the gas migrates through the leak hole and passes into the film of the fluid, bubbles will form and burst. The bursting produces a detectable ultrasound. Leaks with rates as low as 1X10-6 ml/sec have been detected with this method.

Remotely positioned transducers. In some cases, it has been difficult to maneuver and hold a probe at a test position while recording or listening to the generated ultrasounds. Some manufacturers provide multi-directional sensors with a cable that can be positioned in confined spaces. This technique is used to determine the presence of remote leakage without performing the time- consuming procedures required for entering confined space areas.

Valve leak monitoring/trending. An increase in amplitude over a baseline is often a warning signal of impending failure or worsening condition. Valves should be inspected routinely since the information collected can be extremely useful. Aside from go/no-go leak inspection, the worsening condition from acceptable to unacceptable can be determined. For valve monitoring, there should be a consistent test point and conditions within the test object (such as flow rate) should be constant. A baseline should be set and compared to future readings under the same conditions and recording modes

When used in tandem with other technologies such as vibration analysis and infrared thermography, ultrasound has myriad uses. The technology enables knowledgeable inspectors to go beyond the basic applications of leak detection and valve and steam trap inspections, and opens opportunities for improved equipment uptime, energy savings, and safety.

Sealants block leak paths Though leaks of gas or air at a facility are often overlooked, they can become a significant operating cost especially when the situation is chronic. Once a comprehensive survey to detect and pinpoint leaks in a system is completed, the next step is to stop the leaks. State-of-the-art machinery adhesives can reduce costs by eliminating leak paths.

In the average threaded fitting, metal-to-metal contact is approximately 20 percent. Eighty percent is air space surrounding spiral threads, a potential fluid or gas leak path. Loctite, Rocky Hill, CT, supplies engineering adhesives, sealants, and dispensing equipment.

Many situations can cause loosening and/or cracking of fittings, valves, and other connections which result in leaks. Vibration, shock, thermal, and environmental changes all take their toll. Practically all conventional methods of sealing--cork gasketing, pipe dope, or Teflon tape--have their shortcomings.

Conventional gasketing products like cork, paper, and rubber have a tendency to set even when they are properly torqued. When the bolts relax there may be a minute separation leak path. Having an inventory of all size gaskets is virtually impossible, and gaskets can shrink, tear, or deteriorate before use. Also, cutting gaskets is time consuming.

Pipe dope also is no guarantee against leakage. Pipe dope relies on solvents to carry them and form solid seals. When the solvent evaporates, the product dries to form a tough seal. The rigid, brittle nature of pipe dope causes cracking which creates leak paths. And with pipe dope, disassembly can be difficult.

Teflon tape, originally designed as a thread lubricant and not a sealant, can cold-flow out of the pipe and leak. It also can permit overtightening, a situation that may result in threads that lathe up on each other thus increasing the leak path. Another disadvantage of Teflon tape is that it has a tendency to contaminate systems. If a Teflon shred enters a system, it can foul a check valve or other critical component.

The best sealants are based on anaerobic technology. They are a liquid or paste plastic monomer that changes from a liquid to a solid when it comes in contact with metal and when air is excluded.

Because these anaerobic sealants do not dry out but cure without shrinkage, they are excellent when applied to threaded fittings. These sealants provide correct sealing without cold-flow and offer ongoing lubricity that acts as a mild threadlocker. They are also noncontaminating.

The company's anaerobic gasket product for use on rigid flanges allows the flanges to be taken down virtually metal to metal. The plastic gasket material uses similar chemistry to fill in all the voids. Seen under a microscope, these voids appear as mountain peaks and valleys. The sealant fills in the voids between the mountain peaks with a liquid or paste that changes to a solid. And the piece of equipment still can easily be disassembled and removed.

One application of these super sealants is sealing and casting of porosities. A liquid anaerobic sealer can be painted on a clean surface. It will penetrate into every porosity and seal it. For extremely large vats of up to 100 gallons, porous metal parts can be submerged to both seal them and to increase their machinability.

The application of anaerobic sealers is a relatively simple process. And for anyone who needs guidance, some manufacturers conduct in-plant training sessions on the proper selection and application of its sealants.

Once leaks are identified, using gaugeable tube fittings reduces problems, improves equipment reliability, and conserves energy. Drawing courtesy Swagelok.

Gaugeable tube fittings improve reliability The installation of high-end tube fittings and valves often has dramatic results. An energy survey conducted for a pulp and paper company revealed 23 percent leakage in its fluid system. Once gaugeable fittings were put in, the leaks dropped to zero.

Swagelok, Solon, OH, provides connectors for fluid systems ranging from 1/16 in. to 2 in. o.d. Its tube connectors and valves are used in air systems, condensate return systems, hydraulics, pneumatics, analytical instrumentation, acid systems, caustic systems, and small bore process applications.

The company works with maintenance engineers to conduct energy surveys of their facilities. All fittings in a given area of a plant where gas (not liquid) service is common are tested for leaks. Once leaks are identified, the use of gaugeable, two-ferrule tube fittings to reduce problems, improve equipment reliability, and conserve energy is demonstrated.

In a typical scenario, a company representative together with plant personnel checks for air leaks in a compressed air system. Working from as many as 1000 check points, about 24-30 percent leakage is usually identified. This statistic is applied to the company's cost per kilowatt hour and losses are determined. A performance contract to correct the problems is generated. Studies show that properly installed fittings correct leakage to less than 3 percent.

To ensure reliable performance, a tube fitting composed of four components--nut, back ferrule, front ferrule and body--is recommended. The consistency and quality of matched components permit their use in many difficult services. These fittings become a 5-piece connection when affixed to the tubing

The 2-ferrule design and sequential action of the fitting overcome variations in tube material, wall thickness, and material hardness to ensure safe, reliable, and leak-free connections.

Unlike a bite-type fitting which can cut into the tubing and result in a weak point that occasionally may vibrate and break off, a four-piece tube fitting is gaugeable, i.e., every quarter turn is about a 0.0125-in. movement. Consequently there is a go/no-go gauge that enables the person assembling the fittings to put them together and then gauge each one individually during the first makeup to ensure that every fitting will measure up to its properly installed performance factor (1¼ turns).

Leaks may also result from faulty valves, or more commonly from valves whose sealing and packing mechanisms are subject to wear or unsuitable applications. The challenge is to determine the specific application of each valve and choose the right valve for the job.

There are many different types of valves including shut-off valves, regulating valves, uni-directional valves, and pop-off relief valves available for a variety of applications.

Some valves can be pneumatically operated; some may be electrically operated, and some work manually. How frequently should valves be monitored? It is a good idea to check the valve packing and make adjustments periodically according to the cycle requirements of the valve. MT

In 1986, the preventive maintenance (PM) program at Cryovac's Simpsonville, SC, plant was a good one and was considered by some to be the best in the company. The company is the world's largest supplier of plastic packaging and this plant produces blown film.

But we found ourselves asking if the program was good enough for the long haul. What worried us most was that it relied heavily on the extensive experience of craftsmen who had been with the company since the late 1950s. They had installed most of the equipment they now maintained. Many of the most experienced craftsmen were eligible for retirement and very little of their vast hands-on equipment knowledge had been documented.

It was decided that if the PM program was going to continue to meet our needs, we had to document as much of this equipment knowledge as possible while we still had the resources to do so. We felt that a computerized maintenance management system (CMMS) was the best way to do that.

In 1987, we selected a CMMS to get us where we wanted to go. The next question we faced was how to get people who are distrustful of computers to share their equipment savvy with us. After much deliberation, we opted to use the direct approach and simply asked the craftsmen for their help in preserving what they had learned. It was no surprise that craftsmen did not line up in droves to share what they knew. But we were able to get enough assistance to start the ball rolling. With a great deal of help from the maintenance planners, we wrote limited scope job plans and built a three-level equipment hierarchy.

For the next 3 years, the maintenance planners were the only CMMS users. As a result, they were spending all their time converting paper work requests into electronic copies and trying to meet the increasing demand for equipment repair data reports. They were not doing a lot of planning.

It was time for the craftsmen to start inputting repair information and allow the planners to get back to planning.

The first obstacle to overcome was training. It was suggested that initially we should train a core group of craftsmen who would, in turn, help their peers overcome any fear of computers. After training several willing craftsmen, we placed three computer workstations on the shop floor with no instructions about how they were to be used. Craftsmen were able to experiment with the computers with the help of the core group, and felt no pressure while doing so.

This approach gave us the results we were hoping for: craftsmen started requesting computer training. Within 3 months, we were training groups of craftsmen in a classroom environment.

Once the craftspeople were computer literate, we were confident that we now had a foundation to start building an even stronger PM program. Craftsmen were assisting in the writing of job plans. They were inputting equipment failure information and were querying the CMMS database for repair information. The time had come to develop a new written PM plan and put it into practice. The old plan had become too confining.

Technical organization
In our organization, the engineering, predictive maintenance (PdM), maintenance, and technical support groups all report to the same technical manager. It was not always this way but we learned that by having all technical people under the same umbrella, resources and efforts could be optimized. The remarkable thing about combining these groups was that within 2 years after doing so, our goals also merged.

The basic structure of our maintenance organization is such that all maintenance personnel perform PMs although we do have small crews that specialize in preventive maintenance. The PM crews perform PMs on a schedule based on scheduled downtime. The routine maintenance crews (reactive) perform running PMs and augment the proactive crews during scheduled downtime. We feel that it is important to have PM specialists to carry the torch, otherwise the tendency is to fight fires.

Six concepts of an effective PM plan
One of the first lessons learned after formulating our PM plan and putting it into practice was the plan must not be written in stone but instead be an evergreen document capable of change and growth. After several modifications to the plan, we feel that an effective PM plan is based on six concepts:

• Input
• Planning
• Execution
• Feedback
• Documentation
• Accountability

We did not invent these concepts, but we find them to be invaluable tools when organizing a scheduled PM

Where does the input come from?
Input is solicited from any group that has a stake in the results of a scheduled PM. It should be understood that we also use scheduled downtime to make modifications, product changes, and safety upgrades.

Examples of inputting groups are product scheduling, production personnel, reactive maintenance, engineering personnel, predictive maintenance, maintenance support, contractors, and maintenance control (planners).

The planning phase. Planning for the next PM on any piece of equipment actually starts before the current PM on that equipment is complete. Inspections, measurements, and reliability decisions are made and work requests written for the next PM. It is important that these items be documented before they are lost. This will be discussed more in the documentation portion of a PM plan.

All personnel are encouraged to write work requests between PMs and route them to the maintenance planner, regardless of how trivial they may seem. We want to know about the squeaks and squeals. They may seem unimportant, but they may be symptoms of a larger problem. Who better than the people who operate and service the equipment to let us know when it is not well?

Each Wednesday at 1 p.m., the maintenance planner conducts a pre-PM meeting to discuss the upcoming week's PMs with everyone who will be involved in them. The meeting has an agenda and a strict time limit of 30 min. Anyone who has work to be done during a PM must come prepared to talk about what he will be doing and the logistics of the job. After everyone speaks, timelines are established. Start and completion times are set, and assets and resources are assigned by the maintenance planner. Most importantly, commitments are secured from all participants. At the end of each meeting, the PM supervisor conducts a critique of the prior week's PMs.

A lot of communication takes place within a short period of time but if organized correctly, 30 min is ample time. We normally address at least four major PMs (scheduled downtime) at each meeting and it seldom takes more than 20 min total, including the critique. When we first started using the pre-PM meetings, some people came unprepared and many of them walked out of the meeting with hurt feelings because they were cut off at the time limit. It did not take long before people started using the maintenance planner to help them plan their work and get the logistics worked out prior to the meeting, a very valuable PM tool.

PM execution. The plans have been made, the materials purchased, the clock is ticking. The PM supervisor conducts a 5-min communications meeting with the participants prior to the PM to insure that everyone knows his role.

During the actual PM, the three highest priorities are safety, safety, and safety (in that order). The PM supervisor monitors the various groups to in- sure that the plan is being followed. Any changes to the plan are brought to his attention. A decision is made then and there as to how the change will be handled.

Once the PM has been completed, the PM supervisor turns the equipment back over to production for startup.

Feedback following the PM. We do all the usual things: review completed PM work requests, monitor equipment performance, and spot check completed work. But unless that information gets back to the people who did the work, it is only paper. People like feedback; they have a natural need to know how they are performing. Everyone wants to be able to go home, sit at the dinner table with his family, and tell (even brag to) them that what he does is worth doing.

A written critique of the prior week's PMs is distributed each week with the upcoming schedule. The critique is used as a vehicle to apprise the various groups involved in the PM of where mistakes were made, and also serves to announce the successes. The critique gives everyone an opportunity to optimize the PM. When we first started using this technique, there was a tendency to shoot the messenger but that passed with time and understanding.

The most important aspect of our PM program is giving craftsmen feedback concerning the effectiveness of their actions during the PM. This is especially important in the case of PdM work requests. When issued, these work requests are normally accompanied by an infrared scan, a vibration analysis graph, or some other representation of an equipment component in the failure mode. The PM craftsman takes action to correct the problem; the PdM craftsman rechecks these items following (sometimes during) the PM. The PdM craftsman then issues a before and after picture or graph. Craftsmen like to know that what they did made a difference and timely feedback reinforces that need.

We also have a maintenance newsletter that is published quarterly, more often if needed, to inform craftsmen of equipment changes that have taken place and why. These changes are most often reliability motivated and the result of a Root Cause Analysis (RCA) group's work. Craftsmen act as a resource for the engineer-led RCA groups. It is important that people see their ideas in print and their names spelled correctly. Today's software programs make a newsletter very quick and easy to publish.

Constructive feedback given to the individual craftsman will do more to promote a successful reliability-based PM program than anything else I can think of, and it's free

PM documentation. A good CMMS is the heart of an effective PM program. It releases the planner and PM supervisor from the mundane tasks that come with scheduling and planning. An ancient Chinese proverb reminds us that the palest ink lasts far longer than the brightest memory or, in modern terms, if you didn't write it, it didn't happen. When we rely on someone to remember what we are supposed to do on a regular basis, we are going to be disappointed. The computer is another tool to make our life easier, no different than the hand tools we use every day. Why not use it? .

All craftsmen in our maintenance organization are required to input repair and PM task data into the CMMS as the work is completed or at least within the same shift. The PM supervisor reviews the completed work requests daily to assure accuracy. If equipment PM or repair data is unclear or incomplete, the work request is returned to the craftsman for additional input. This data will later be used to make reliability decisions. Reliability is the prime goal of any good PM program.

Parts and materials are requested through CMMS-generated work orders. This gives all craftsmen (and others) access to the maintenance planner without crowding him out of his office. Requiring craftsmen to document equipment conditions within computer-generated work requests insures that the parts are ordered and affords the craftsman a way to follow that process. When the parts arrive, the planner notifies the craftsman's supervisor through the CMMS. The work then can be scheduled.

Many PM tasks are not scheduled on a monthly basis but at infrequent intervals. A CMMS is ideal for making sure those tasks do not get lost in the shuffle. For example, some pressure vessels, by law, require inspection every 15 years. Special work instructions can be documented and issued for tasks that people do infrequently. A CMMS also gives the supervisor or planner the flexibility to add or delete tasks if conditions or situations change. These task modifications can be done in a matter of minutes.

The bottom line is that all maintenance tasks, scheduling, parts procurement, and labor are documented. This may sound like a lot of documentation, but when spread among dozens of people it is much easier than the supervisor or planner trying to keep up with all of it. With each piece of documentation, equipment history is preserved. If you do not have a history, how can you possibly plan a future?

One note of caution: Our mission is reliability-based maintenance; try to avoid the data stream manipulation in which people get caught up. Leave the data massaging to the bean counters. Get the data that you need to improve the process, institute a reliable fix, and move on to other things that are more deserving of your attention.

Accountability. Accountability is simply the process of closing the loop. If a reporting system is open ended, it is doomed to failure. There has to be a vehicle to assure that tasks have been completed, parts are on order, equipment is scheduled, and people are paid. The CMMS is that vehicle.

The CMMS is a very effective tool for closing that loop. The CMMS is simply a maintenance tool and should not be mistaken for anything else; it does not control anything, and should be viewed as a reservoir of information. It allows all maintenance personnel access to the information that is needed to accomplish their missions effectively. It does not take control away from the PM supervisor; it gives him expanded options.

Expectations of a well-executed program Over the past 6 years, we have seen our CMMS coupled with PdM technology and the result has been a very strong reliability-centered maintenance (RCM) program. When the entire maintenance organization with the support of its production counterparts is com mitted to putting this technology into practice, the results are most satisfying.

Over an 18-month period, unscheduled downtime has been reduced to 47 percent of its base line level.

An effective PM program must have a plan that is workable yet flexible enough to meet equipment needs. Craftsmen should understand and be trained to execute the plan. All efforts should be concentrated on tasks that add value. It is very important that everyone understands and appreciates the company's business.

Only the people who participate within its set boundaries limit a PM program. Those willing to push those boundaries will be rewarded with equipment that is more reliable and a knowledge that allows them to predict how that equipment will perform under any circumstance. Once the tools are in place, the people that use them must be committed to the process. Imagination is the only limiting factor.

I would offer a word of caution for those that cannot stand pain. Keep doing it the way you have always done it and nothing will change. But if you can tolerate the pain of change, the rewards are worth it. MT

Charlie Harrell, a PM supervisor at Cryovac, a division of Sealed Air Corp., has worked in a maintenance environment for 36 years. He can be reached at P.O. Box 338, Simpsonville, SC 29681-0338; (864) 967-1480; e-mail Charlie.E.Harrell@Sealedair.com

The goal of any company's information systems (IS) department is to serve the business. When the typical internal customers are asked about the level of service they receive from their IS organization, the response is often not favorable. IS organizations are usually working very hard with limited resources and competing demands, and they often lack clear direction and priorities. With this perception so common, and with outsourcing on the rise, most IS organizations must evaluate their service and ensure they understand and are meeting the needs of their customers.

Substitute R&M for IS in the previous paragraph and it covers issues facing many reliability and maintenance organizations. The following discussion about recreating the IS organization provides insight for building a better working relationship with IS in your company. It also contains ideas for improving the effectiveness of the group responsible for managing maintenance information in your department, as well as some suggestions on how you can improve your relationships with internal customers such as the production organization

To develop into a highly reliable and functional service organization, the department must follow four basic steps. First, the group must assess current practices to understand the state of the organization. Next, the group must build a foundation for change. Only by developing a clear vision of the ideal state can change be successfully implemented. Third, the plan developed must be executed and implemented within the organization. Finally, as the benefits of the improvements are realized, the values and ideals generated must be reinforced in order to create an organization focused on continuous improvement.

Understanding the past
To better understand the current state of the IS organization, it is valuable to look at how the group may have evolved. This provides perspective and facilitates learning from the past without having to recreate it. Many IS teams were formed out of immediate necessity rather than by design. As companies grew to depend on technology more and more, they realized that a dedicated group of people was required to manage it. As the responsibilities of these groups grew, processes and procedures were developed by trial and error. Once the team found a procedure that worked, it became the only way to do it. Hence the phrase, we have always done it this way. When the procedures did change, it was usually because of a customer complaint or a new technology demand. These technology demands frequently came from customers in the company who believed they needed the latest and greatest tool or from the manufacturer that discontinued support for the current tool. Once this oc- curred, the small group of users that was upgraded to a new system with new software be- came incompatible with everyone else.

Once this cycle of change begins, the company is left with the choice of living with an inconsistent computing infrastructure or interrupting the business to upgrade systems and software for everyone. Both of these choices are painful. Most organizations, by default, live with the inconsistent computing infrastructure. As a result, the IS group is always fighting the latest fire rather than working on the larger picture of infrastructure, standard operating procedures, and documentation. This makes moving forward extremely difficult.

As this situation develops into something unmanageable, the frequent comment is that something should be done. That may be true, but what specifically should be done? That is a question only answered by an objective assessment and analysis of the company and the IS organization.

Assessing the current state
The assessment of the IS organization will provide information on how the department is serving its customers. To do this, the assessment must focus on the processes used rather than the technical elements of the infrastructure. Although a technical element such as the choice of computing platform is important, customer service is usually based on how well the customer's needs and expectations are met. Elements such as project management, work planning and scheduling, communication, and team culture are important to the customer. Good project management and work planning and scheduling set expectations for the team while clear communication allows the customer to understand and be a part of those expectations. The team culture is an indication of the overall attitude that the IS team member exudes to each client, greatly affecting customer satisfaction.

This assessment should be as objective as possible. This can be accomplished by involving a representative sample of people across the company so that all viewpoints are considered. The discussion should be based on a predetermined set of questions or criteria that the organization wants to measure, led by an unbiased facilitator. An example of points that could be included in the assessment criteria is shown in the accompanying Systems Assessment Grid, which includes a range of process-oriented topics. The criteria will prevent the team from getting off track or allowing personal bias to sway the outcome. Management support of the assessment is key to its success. If management does not support the objective findings, then the participants will not be honest due to a fear of retribution. When the assessment is complete, the results should be presented to the company and made public for all to see. This will encourage open and honest communication during and after the assessment.

Build a foundation for change
Once the IS organization has been objectively assessed, the team can decide what to change. As with any organizational change project, it is important to develop a strong foundation. The basic steps of a successful project foundation are to decide what better looks like and to create a game plan for getting there. These steps are not successful if the manager of the change process simply develops the answers and publishes them. There must be joint prioritization of project needs among the whole IS organization and a representative sample of its customers.

When a cross-functional group, which includes the IS organization, project management, and internal customers, reaches clarity on what should be done, it can proceed to develop a prioritized plan. Once the consensus for the project plan has been reached, this combined group must commit to it. This includes agreement about the necessary tasks, the relative priority of these tasks, the overall timeline for the project, and resource requirements for the change. Without this clarity, consensus, and commitment, the project is destined to struggle during implementation. This up-front planning is often avoided because it takes time and money while customers are demanding immediate action.

In order to develop the detailed improvement plan, the group must have an overall direction to follow. This may include information about the purpose, values, and principles of the organization. Defining such principles may seem difficult, but they should be kept simple. Every organization's principles will be based on their unique characteristics, yet some that should be followed are outlined in the accompanying section Key Principles for Information Systems.

Implement your vision
Once the details of the goal are determined, it is up to good planning and project management to implement the project. This planning and management role is often underestimated; don't be surprised if it becomes a full- time job. Whether an internal or external resource is identified to lead the project, he or she should be dedicated to the management of the project and understand the organizational change process. Implementation is by far the most visible and lengthy part of the project, but it is not the whole project. It is only as good as the assessment and foundation building that preceded it. Without them, the implementation phase is often disorganized and painful.

Reinforce your values
Exercise the values and principles the organization is seeking every day. This means that the company should know and understand the principles and values that were developed during the foundation-building process. The reinforcement of values and principles can come through personal example, process measurement, and a system of accountability. Process measurement includes things such as schedule compliance measurements, backlog trending, and amount of emergencies. Accountability is the combination of performance measurement and performance review. Simply stated, accountability ensures that tasks get completed. The process measurements and system of accountability details also should be developed during foundation building or early in the implementation so that expectations are clearly set.

The information systems and reliability and maintenance organizations can be challenging to improve due to the unique requirements placed on their members. It is important to follow the four steps outlined previously to create a great organization. First, understand what the current practices are. With that state in mind, form a team to define values and provide a vision of success for the organization. With that solid foundation to stand on, the organization is ready to make the real changes necessary to perform the way it envisions. Then, at last, the change must be reinforced for it to become a way of life. When the organization is aware of its history, current state, and future vision, it is ready for the next step, the revolution. MT

David Bair is a senior consultant at Reliability Management Group Minneapolis, MN; telephone (612) 882-8122

Implementing technological advances in a vintage mechanical drive design will result in a more efficient, durable system that can meet increased production demands.

The quality of mechanical drive system components has evolved and improved over the years. Advances in drive system design, manufacturing capabilities, and materials technology allow existing components to be replaced with more durable and efficient equipment with significantly higher power densities.

Technological improvements
Some of the greatest advancements in drive system components have occurred with the girth gear set, consisting of a gear and pinion. Material technology, manufacturing techniques, and rating standards have changed, and now provide uprating and upgrading opportunities to facilities with vintage mechanical drive system designs.

First, material technology for steel castings has been refined, resulting in techniques that have drastically improved quality, integrity, and component hardness. In the 1950s, state-of-the-art in material technology could produce gear blanks with a hardness of only 180 HB (typical) or 225 HB (maximum) and pinions with a hardness of 265 or 285 HB. Now, with the availability of high hardness girth gears and through-hardened or carburized pinions, durability and strength rating increases can be realized with the installation of such components. When installed as a replacement for an older drive system, these latest technology gears and pinions can achieve durability rating increases exceeding 100 percent and strength rating increases exceeding 50 percent.

Second, manufacturing methods have improved component quality. Tooth accuracy of pinions and gears is significantly greater because of modern machinery used in the manufacturing process. The gear tooth quality that 1950s or earlier vintage gear cutting equipment could produce is in the range of AGMA 6 for gears and 8 for pinions, compared with today's available quality levels of AGMA 10 for gears and 12 for pinions.

Third, the equipment rating standards developed by the American Gear Manufacturers Association (AGMA) have changed. In an attempt to quantify difficulties that arose in the manufacture of such large gears, rating methods differed for open girth gears and enclosed gears. Originally, the AGMA reduced the rating of large gearing, such as girth gears. Now, because of improved materials and manufacturing methods, rating standards more accurately model the actual performance of all gears.

Options
Depending on the required increase in production and the budget, there are different uprating and upgrading approaches to achieve the necessary output level: replace pinions, recut girth gears, replace girth gears, uprate the gear drive and couplings, or a combination of these options.

Replace pinion in girth gear set.
Taking into consideration improvements in materials, manufacturing, and rating standards, a higher hardness pinion can be installed to increase the power density and production level of a vintage girth gear set.

AGMA rating practices allow a rating increase by increasing only the hardness of the pinion. The replacement pinion can be either through-hardened or carburized, depending on the rating increase desired.

It is important to note that the rating increase from replacing the pinion assumes the girth gear is in like-new, as-manufactured condition. This is typically not the case and, therefore, the full rating increase may not be realized. An exact value for adjusting the rating of the girth gear set due to gear wear or damage cannot be assessed without a thorough inspection of the gear.

When replacing a girth gear set's pinion, tooth modifications can be performed. Grinding the new pinion's teeth to specific modifications will increase load carrying capacity and operating contact, and extend service life.

For a typical application such as a grinding mill, a significant production increase may be possible by increasing the number of pinion teeth. This results in a lower total gear ratio, which increases the mill's speed and, in turn, production ability. The exact speed increase depends on the original number of pinion teeth. For example, increasing the number of pinion teeth from 19 to 20 will increase the mill's speed by 5.3 percent. A 4.5 to 6.25 percent increase in speed is possible when adding one additional pinion tooth. But an increase in speed will put the driven equipment closer to a system's critical speed. The original equipment manufacturer should be consulted when making this change.

Increasing the number of pinion teeth may require that the girth gear set's center distance also be increased. Allowance for this is usually available in the pillow block foundation bolt holes. If not, the bolt holes can be slotted to accommodate the increased center distance.

When implementing any system upgrade that will increase speed, remember that the motor must be capable of providing the extra power required to drive the system at the increased speed. This is typically not a problem, as most systems draw less than full motor power. If a facility is operating at motor nameplate power, adding additional cooling to the motor can usually increase the motor power. When faced with this situation, the motor manufacturer should be consulted for a proper recommendation.

Replace pinion and recut girth gear.
Recutting the teeth of a girth gear restores the original tooth form and makes it possible to take full advantage of a new pinion (see Fig. 2). In addition, the gear structure is completely inspected during the recutting process and any defects are repaired. This provides the structural integrity required to transmit the increased torque. Defects in the gear structure are identified by nondestructive testing methods, such as magnetic particle and ultrasonic inspection, and weld repaired (see Fig. 3). If required, weld repairs are made before the girth gear is completely stress relieved and/or heat-treated to ensure proper integration of the repair with the base metal (see Fig. 4). All surfaces of the gear are machined to ensure dimensional accuracy and geometric tolerances that meet or exceed the original design.

A complete design review is undertaken to validate all aspects of the gear and pinion design and manufacture. This updates the design with current design methodologies and rating practices.

The new tooth surfaces also yield increased efficiency benefits. A recut gear operating with a new pinion restores the gearing to 99 percent or greater efficiency (see Fig. 5). This can translate into significant operating cost savings. For example, a 1 percent increase in efficiency for a 1000 kW (1341 hp) mill will result in savings of $3500 per year, using an electricity cost of $0.04 per kWh.

Fig.2 Representative tooth damage that can be repaired by recutting the teeth.	Fig. 3 After cleaning, nondestructive testing identifies defects in the gear blank structure that then are removed.
Fig. 4 The defects identified and removed in Fig. 3 have been finish weld repaired. The repaired areas are not visible after the gear blank is painted.	Fig. 5 Major tooth damage has been removed and the tooth-working surface has been restored to like-new condition. The remaining damage is minor and will not affect the operation of the gear.

The cost of refurbishing a girth gear will vary, depending on the amount of repair that is required. Typically, refurbishing an existing girth gear is 30 to 50 percent of the cost of a new gear. The exact final cost heavily depends on the amount of weld repair required.

Replace pinion and girth gear.
A third uprating option is replacing both the pinion and girth gear. This allows the use of the latest design methodologies and rating practices, and takes advantage of the improvements in manufacturing and materials technology.

Replace gear drives and couplings.
Uprating the girth gear and pinion is useless if the main gear drive and/or couplings cannot transmit the increased torque. In some situations, these components will need to be uprated or replaced with new counterparts.

The gear drive can be replaced with a higher hardness gearing. For a gearbox with through-hardened gearing, the typical uprate using only a carburized pinion is 15 percent. Replacing both the pinion and gear with carburized elements will result in an uprate of up to 50 percent.

At the same time the gearing is replaced, new bearings using latest material and manufacturing technology should be installed. Bearing manufacturers have released E-type spherical roller bearings that have significantly more load carrying capacity than similarly sized standard bearings. The typical uprate using E-type bearings in place of standard bearings is 15 percent.

Rating practices for bearings include adjustment factors for lubrication, cleanliness, and load zone. These factors can either increase or decrease the calculated life of a bearing. The result is a much better understanding of the actual operating life of the bearing. New seal design technology such as Taconite, Magnum or noncontact also can be installed during a gear drive upgrade to provide better leak protection, reduced operating temperatures, and longer seal life.

New seals with higher allowable operating temperatures, such as Viton seals, are additional technological improvements that can upgrade and uprate a vintage drive system.

Shaft couplings have experienced a similar uprate over the past 30 years. New materials and manufacturing processes have increased shaft coupling power density 70 percent for grid-type couplings. The rating of gear-type couplings has increased at least 55 percent, size for size. In essence, a dimensionally interchangeable coupling with a much higher rating can be used or a smaller coupling with the same rating can be installed to significantly reduce costs.

There are many ways to increase the power density of existing mechanical drive systems in order to meet increased production demands. Improvement of materials, manufacturing techniques, design methodologies, and rating practices over the past 40 years has resulted in mechanical drive systems that are no longer a limiting factor in achieving higher production goals. MT

Bill Hankes is engineering manager, mill products at The Falk Corporation, P. O. Box 492, Milwaukee, WI 53201-0492; (414) 937-4566; e-mail bhankes@falkcorp.com; Internet www.falkcorp.com

Robert C. Baldwin, CMRP, Editor

Zero value gaps: Products or services custom fitted to each customer to maximize the value each receives. Values, of course, are different for different customers. They want the product or services tailored to their needs, delivered at the appropriate time, and at a price point that makes the deal satisfactory.

Zero learning lags: Management of the entire life cycle of knowledge--from creation to dissemination. It is suggested that the components of a zero learning lag organization include an environment for learning; management of the knowledge in useful chunks; and an infrastructure supporting seamless integration of computing, communication, and content technologies.

Zero management: Every person in the organization has the ability and the permission to do whatever needs to be done in order to produce value for customers. It is implied that people and teams are aligned with the corporate whole.

Zero resistance: A process in which there are no obstacles to performing whatever tasks are required. Zero resistance requires that individuals have achieved personal mastery of tasks and that they are empowered to follow them through.

Zero exclusion: All people and organizations who need to be involved are included--automatically--with neither physical nor technological boundaries to limit accessibility. The Zero Time organization is a proactive organization that anticipates, senses, and responds to the environmental changes influencing completion of the corporation's mission and goals.

Reliability and maintenance practices and technologies fit nicely into these Zero Time buckets:

• Proactive maintenance and RCM advance the cause of zero value gaps
• Condition monitoring and computerized maintenance management systems facilitate zero learning lags
• Autonomous maintenance is the essence of zero management
• TPM is based on zero exclusion
• Planned maintenance promotes zero resistance.

What time is it in your maintenance operation? MT

Symptom: Your CMMS has become too slow.
This can occur for several reasons that have nothing to do with your CMMS software. First, when was the last time you archived old work orders? If your system is bogged down with 75,000 old work orders, a software upgrade is not going to help. Most reputable CMMS provide for the archiving of old work orders.

Also, what has changed recently? Is your server now handling more PCs than before; are there other applications on the network that have been added that might impact the speed of your network or your server? One of our clients installed a DOS-based application on their non-dedicated server. Every time they fired up their old spreadsheet software, the speed of the CMMS dropped by 50 percent. They were convinced it was the CMMS until shown otherwise.

Also, is your CMMS database engine being used by other software applications? Sometimes you're not just sharing your server, you're sharing your database as well. If it's tuned to work well with one application, it can have an impact on your CMMS. You also can suffer performance issues if the other software is causing a significant number of hits on the engine, cutting down on the amount of processing time left for you.

Symptom: Your CMMS doesn't have features that others do.
CMMS vendors are always adding new bells and whistles, and sometimes people feel behind the times with their current application. That's a typical feeling and may very well be founded; however, there are a few little things that you should check first.

Have you been keeping your current CMMS up to date? Almost all vendors offer a maintenance program for their software to ensure that you get bug fixes, new version of the software (with--surprise, new features!), etc. However, what often happens is that in budget cutting somewhere along the lifespan of the software, someone decides not to renew the maintenance on the CMMS. It seemed like a good idea at the time, but can leave you with an out-of-date application.

What about re-training your staff on the software? Most CMMS training is done in the same manner a bad gene is passed on--it is done by an employee (now long gone and retired in Barbados) who trained a handful of people, who trained a handful of people, and so on. Each generation of training lost more and more of the skills until they sometimes reach the point where the staff is convinced the CMMS is nothing more than a work order engine. Sometimes when you're convinced your CMMS can't do something, you need to crack open the manual to find out if it's possible in the first place.

Also, are the new features you've read about in brochures, etc., really going to add to your maintenance department's productivity, or do they just look good? This is the old adage of being the sizzle, not the steak. You get the idea that the new functionality will be a big help, when in reality, it's not practical for your operation.

Symptom: Your CMMS doesn't work with other company applications.
The CMMS as an enterprise-wide application is not a new concept, but often not implemented correctly. Also, the maintenance department often is on the low-pecking order for new software, meaning that the accounting, purchasing, or enterprise management software your CMMS used to connect with has been upgraded or changed while your CMMS remains the same. That, or you never attempted to integrate your CMMS to any other application(s).

First and foremost, check with your current CMMS vendor and see if they allow or can provide integration to whatever software you want to link up with. More often than not, the larger vendors are more than willing to do so if you can specify what that linkage needs to be and how it will work.

Keeping fasteners tight, particularly threaded fasteners, seems like a simple task, but the moving nature of the machinery they are used on is what makes them so troublesome. How nails and rivets work is fairly well known. But because the physics of the threaded fastener is not as well understood, it tends to cause the most problems.

Causes of loosening
Threaded fasteners are employed primarily to clamp objects together using tension. Rotary force or torque imposed on the fastener provides that tension. Problems occur when this clamping load deteriorates.

About 85 percent of the torque and effort of tightening a bolt is absorbed by the friction in the threads and under the head. Only 15 percent produces clamping load. Therefore, high torque may be absorbed by high friction and not produce tension. Torque is not the most precise method of controlling clamping load, although it is the most common. When bolt and nut manufacturing is closely controlled, the tension produced in a bolt for a given torque varies up or down by 15 percent.

Although it is always the first suspect in any case of lost clamp, vibration, as commonly perceived and observed, is not capable of bolt loosening by itself. If vibration is violent enough to cause shifting of the threads, then it will cause loosening, and in only 50 to 100 cycles. However, vibration that violent is usually perceived as shock, shudder, or impact. Toward the end of the loosening cycle, common vibration can and will rattle the fastener loose. This is why it often takes full blame for loosening.

The actual cause of loosening is side-sliding or shifting of the threads. The empty space between the threads of a nut and bolt leaves room for movement that leads to self-loosening and loss of clamping force. The friction in the threads and under the head of the bolt is reduced to zero when the clamped parts and threads slide sideways to the bolt axis.

Each time this happens, the bolt can unwind by itself. The loosening process of a non-locking fastener starts with the first motion. It normally takes less than 100 side motions to completely loosen a bolt.

This shifting can occur any time the side force exceeds the friction between surfaces, as produced by the clamping load. There are three common causes of shifting:

• Bending. Bending of parts causes stress on the friction surface. If slip occurs, the threads and head also will slip. Each slip causes a partial downhill or unwinding slip in the threads. After 50 to 100 of these, the bolt is completely loose.
• Thermal expansion. Differences in temperature or in clamped materials can cause the same effect as bending. If the effect is strong enough to cause side-slip, then downhill slip also will occur and loosening will result.
• Applied loads. The impact of loads applied directly to the fastening point can cause side-slip as well.

Any one or combination of these conditions can occur from shipping trauma, extreme heat or cold, or just plain abuse. The effects are cumulative and self-accelerating. As these affect clamping load, there is increased probability that side-sliding will occur.

Threadlocking
Various methods and devices have been employed over the years to reduce or prevent loss of clamping load in threaded fasteners.

The earliest attempts involved the use of lock wires and split pins in conjunction with nuts and bolts with holes drilled in them. Although effective, these measures had some serious drawbacks. Each fastener had to be the correct length, and the holes had to be aligned on each individual bolt. Consider the difficulty and time required using this method to assemble numerous parts requiring many threaded fasteners.

As fastener manufacturing skills improved, more complex methods of threadlocking were developed. Two of the most common mechanical methods of threadlocking are thread distortion and the use of washers. Although these methods can be effective for short-term threadlocking, anaerobic threadlockers can provide short-term, long-term, and even permanent tightening when necessary.

Liquid threadlockers
The first chemical threadlockers, developed by Loctite, eliminated many of the design faults and shortcomings of threaded fasteners. Chemical threadlockers are anaerobic liquids that cure to a tough, solid state when activated by a combination of contact with metal, and a lack of air. The resulting cured material is a thermoset plastic that cannot be liquefied by heating, and resists most solvents.

The purpose of threadlockers is to lock and sometimes seal threaded components without changing fastener characteristics or altering torque-tension relationships. In addition, chemical fasteners offer a number of other advantages over mechanical tightening methods:

• Breakloose and prevailing torque. Liquid threadlockers find their way into tiny imperfections of threads. As they cure, these imperfections serve as molds for thousands of tiny keys that resist fastener movement in any dimension.
• Anti-corrosion. Because threadlockers fill the voids between threads, they block the entry of moisture, preventing corrosion and subsequent seizure.
• Strength control. Most threadlockers are graded by their various strengths and characteristics into distinct classifications. The different formulations of Loctite threadlockers, for instance, are distinguished by the color of the threadlocking material: low-strength is purple, removable is blue, permanent is red, and the penetrating formula is green.
• One size fits all. Because they are liquids, threadlockers do not come in different sizes. The same bottle that locks in a tiny screw also can be used on a large bolt. Stocks of various size mechanical threadlockers are no longer necessary.

Selecting the right threadlocker There are several key factors to consider when choosing a threadlocking compound:

1. Shear strength. If all threaded fasteners were designed never to be removed, then only one type of threadlocking compound would be necessary, the strongest available. Most assemblies that are held together with threaded fasteners will, with varying frequency, need to be dismantled for repairs, maintenance, or adjustments. Consequently, threadlockers of various shear strengths are available.
2. Cure speed. The cure speed of threadlockers can vary, depending on several factors, including temperature, base metal, surface treatments, clearance between parts, and surface cleanliness. The use of chemical primers can speed cure and result in higher ultimate strength.
3. Gap filling requirements. Most threaded fasteners are designed with some clearance between their mating surfaces. Larger clearances between mating surfaces require more product to fill them. Thixotropic liquid threadlockers will easily fill clearances in threaded fasteners, without migrating to other areas of the assembly. Where a higher shear strength product is required, and product migration is considered a potential problem, a higher viscosity compound is recommended.
4. Operating environment. Both chemical resistance and operating temperature should be considered when selecting a liquid threadlocker.

The chemical resistance properties of threadlocking compounds vary between different grades. The most popular anaerobic products will generally resist water, natural or synthetic lubricating oils, fuels, organic solvents, and refrigerants.

Like most organic materials, threadlockers lose strength at elevated temperatures. Most show significant strength retention at temperatures up to 300 F (150 C). Hot-strength formulations can increase this working temperature to 450 F (230 C) for those applications where it is considered necessary.

Removability
The most common myth about liquid threadlockers is that once they are cured, they cannot be removed. In fact, all threadlocked fasteners can be removed. Different grades of threadlocker can be used depending on the task. Fasteners secured with low- and medium-strength grades can be removed with common hand tools. Those secured with high-strength grades can be removed by applying heat for a specified time.

Threadlockers are not just for specialized uses, either. They perform effectively on fasteners and threaded assemblies of any type and size, in any kind of equipment. MT

Information supplied by Robert A. Valitsky, a manager of technical communications at Loctite Corp.'s North American Engineering Center, 1001 Trout Brook Crossing, Rocky Hill, CT 06067; (860) 571-5416; Internet www.loctite.com

There is universal agreement that improved machine performance can control and reduce manufacturing costs. That was the goal of Coors Brewing Co., Golden, CO, when it commissioned a maintenance benchmarking study of its can and bottle packaging lines and the warehouse operation.

The study results compared the maintenance operations at Coors to a world class model and other companies in similar industries. The benchmarking process, developed and administered by Charles Brooks Associates, Inc., uses a grading system to determine a company's level of maintenance effectiveness.

Although the brewery's maintenance operations performed better than most of the comparison companies, two major opportunity areas were identified: planned maintenance and mechanic training.

As a result, Gene Rowe, senior consultant, and Coby Frampton, partner and president of Charles Brooks Associates; and Paul Altimier, director of can packaging at Coors, designed a program for the company that later became known as Reliable Predictable Manufacturing (RPM). The goal of RPM is to improve equipment performance and control or reduce manufacturing costs.

The RPM process includes many of the elements of other improvement processes such as critical component analysis, equipment upgrading, planned maintenance, and performance evaluation. It differentiates itself through the use of Defined Equipment Standards (DES) as the basis for maintaining and operating equipment, as well as being the vehicle for achieving employee participation, skills enhancement, and production/maintenance cooperation.

The 20 steps required to implement RPM are shown in the table "Steps for RPM Implementation." The first eight have to do with proper planning and setting the stage for change. The last 12 steps outline the implementation of the DES, modification of preventive maintenance routines and audits, and mechanic training. While quite specific in application, DES is essential in implementing Total Productive Maintenance (TPM) and Reliability Centered Maintenance (RCM).

To insure that the project goals were being met, a baseline was established measuring unscheduled downtime, quality, and production.

Monika Seiler, the plant's RPM process manager, and Rowe involved the hourly mechanics and electricians from the onset to gain support for the RPM process. They developed the DES, prepared updated preventive maintenance (PM) routines, and conducted extensive, formal peer-level training. They also made presentations to upper management explaining their work and the benefits that were accruing. Hourly maintenance personnel took ownership of the analysis of line performance data and developed specific action plans to correct recurring problems.

Understanding how the program works and how it will affect them was essential in gaining their confidence and cooperation. The mechanics and electricians were empowered by the knowledge that they had an opportunity to participate in and assist with the development of a key business strategic program that would affect the company's bottom line. In order to request program funding, the supervisors, mechanics, and their appointed peer leaders met with L. Don Brown, senior vice president, operations and technology, to voice their support of and commitment to the program.

In order to determine the pilot production line, the team analyzed downtime data and identified the line with the greatest problems.

The variables analyzed were quality, production data, unscheduled downtime, and maintenance spending. Can Line No. 10, it turned out, produced the most can defects and received the most maintenance dollars.

Can Line No. 10 was broken down into major equipment systems or subassemblies. An analysis of subassembly downtime and maintenance cost determined which subassemblies were causing the can defects.

Next, current machine settings were identified, documented, and evaluated with input from mechanics from all shifts. Not unusual with many companies, each mechanic used a different setting. After gaining consensus from the mechanics, machine settings were adjusted to a set standard and were documented. This standard setting became part of the DES.

Once the can line's collator was upgraded and set to the new machine standard, can defects were reduced by 87 percent. After further evaluation, a variable speed drive was added to reduce pressure from the accumulator table, thereby eliminating all defects that prevented shipments.

Machine performance was monitored at the set standard and adjustments were made as necessary to optimize machine and line performance. When optimal machine and line performance were achieved, the machine settings were documented in the DES. Maintenance personnel took digital pictures of each machine and documented how the settings were achieved.

The use of digital pictures proved to be an important training tool as well as a creative way to document the DES for all the line equipment.

As a result, PM procedures were modified to reflect changes in the machine settings and new PM procedures were developed. Machine audits also were developed and implemented to assess the level of maintenance received and the current machine condition to ensure optimal machine performance.

Detailed PM procedures were developed once the optimal level of PM was achieved, and entered into the computerized maintenance management system.

At this point, the training process was begun to train mechanics on the new machine settings and PM procedures. A technical job skill training process called the Analytical Method of Training (AMT), used by Charles Brooks Associates since 1971, was implemented.

It begins with a thorough analysis of training needs, followed by standardization on the best trainable method. Once the method is known, skill development begins with peer-level instructors providing training until the trainee can perform at the expected level. When single-cycle skill has been achieved, the stamina buildup phase begins. The entire process is measured and monitored through extensive testing and performance demonstration.

Through an improved planned maintenance process starting with Defined Equipment Standards and mechanic training, reduced maintenance costs and improved efficiency can be achieved. The involvement and assumed ownership of the program by hourly employees, supervisors, and managers assures long-term results can be realized. MT

Katherine Berntzen and Gene Rowe are consultants with Charles Brooks Associates, Inc., P. O. Box 11758, Charlotte, NC 28220-1758; (800) 868-3553; e-mail rtt@charlesbrooks.com

Recent discussions with industry leaders make it clear that a process for equipment reliability management is emerging. The process seems industry independent and includes several generic principles. Those out in front of this movement recognize the necessity to express reliability issues in financial terms.

Interest, involvement, and drive from the top are characteristic of leaders. Leaders value people and recognize that people, not things, create value. They recognize that maintenance is not a stand-alone, stovepipe we run it, you fix it process but rather a vital, inseparable part of manufacturing. Some call it capacity management, others TPM. In truth, it is both and much more.

There is a growing recognition that equipment effectiveness, maintenance and reliability improvement, and the heavily promoted TLA (three-letter acronym) solutions are worthless unless they can conclusively demonstrate improvements in availability, production output, quality, and conversion cost. If you can't prove it, you don't get it--or, if you have it, it might be taken away.

Leaders identify deficiencies that detract from production objectives by occurrence, cost, and lost opportunity. The information leads to a prioritized action plan. Those following this process recognize that the mix and priority of deficiencies shift with changes in market conditions and completion of improvement initiatives. We're dealing with a moving target.

Several readers commented that corporate downsizing and managerial indifference had eroded competency in such vital areas as predictive monitoring, precision alignment, and balancing. I'll bet those indifferent managers's eyes would light up if we talk financial numbers.

Let's begin with the big picture. Authoritative sources report that North American manufacturing spends between $200 billion and half a trillion dollars a year on maintenance. An article in this magazine a couple of years ago asserted that over 60 percent of maintenance is preventable. A reduction of just 15 percent represents a minimum of $30 billion. Most industry leaders are going for 25 percent minimum--$50 billion at the lower estimate of maintenance expenditures.

With this huge opportunity, the big puzzle is why isn't the maintenance optimization industry humming along? Paraphrasing Pogo, the swamp sage, perhaps the enemy is us.

I suggest both the opportunity and means are totally within our control. Those who are complaining about not being listened to, and having their efforts hampered by people with the interest and attention span of a five-year-old, should think back to their first difficult problem. Most technical experts developed their expertise with interest, enthusiasm, curiosity, education, time, and observing a lot of perplexing problems up front and personal. We're now finding that technical expertise and knowledge that we're doing things right isn't enough. The value of our efforts must be proven in thousands of dollars.

What are the barriers that keep reliability experts from learning the skills necessary to connect technical results to improvements in production effectiveness and business performance? I suggest the conditions are exactly the same as when today's reliability experts viewed their first frequency spectrum, balanced their first fan, repaired their first pump, performed their first shaft alignment. Today there is an advantage--recognition that technical knowledge must be reinforced with compelling financial justification demonstrating value. Think of your first complex problem. Everyone recalls being confused, uncertain, and more than a bit inadequate. What did you do? You learned and added to your knowledge. That's called experience.

Today we're faced with another challenge. Have you asked yourself what are the top five concerns of your plant manager? More importantly, how can you contribute to the manager's success and security in those areas? Find out, learn how, gather experience, and increase your personal satisfaction and worth. MT

Maintain Assets Network (MAN) has served as a vehicle to substantially strengthen the culture of maintenance and reliability in Amoco's chemical businesses over the past two years. It is one of five "networks of excellence" established by the Amoco Chemicals Manufacturing Council in 1996 to drive improvement in 20 manufacturing metrics established to measure performance of the company's chemical sector and chemical plants.

MAN met for the first time in late 1996 and consisted of representatives from nine U.S. chemical plants, three non-U.S. chemical plants, and one representative for eight fabrics and fibers plants. The group met bimonthly and began to set priorities, with initial activities centered on improving three metrics:

1. Maintenance costs, as a percent of estimated replacement value (ERV)
2. Availability ratio (reliability)
3. Sustaining capital

Baseline costs were established and a goal set for 1999 to reduce maintenance costs by 50 percent. The group struggled with the magnitude of the goal, differing maintenance accounting practices, and the calculated replacement value of the plants. One network member was assigned the task of developing standard guidelines for maintenance cost accounting and replacement value calculations. MAN members also devoted time to understand each other's organizations, work processes, current improvement activities, and opportunities.

Strategy development
Early in 1997, it became apparent that the group had to move beyond discussing the merits of the metrics and start to impact them. The network established five subcommittees to cover:

1. Long-term MAN strategy
2. Reliability improvement
3. Pumps
4. Centrifuge
5. Product change

The first subcommittee mission was to develop long-term MAN strategies to give the group needed direction. The other four subcommittees were to research and recommend best practices in their assigned areas that could be implemented at the plants to quickly improve equipment reliability and to reduce maintenance costs. They covered equipment reliability, pump maintenance and reliability, centrifuge maintenance and reliability, and reducing lost capacity from product changes.

It was believed that all of these efforts, if embraced by the plants, would deliver the overall asset effectiveness (OAE) goal of a 25 percent increase and the maintenance cost goal of a 50 percent reduction. The MAN strategy is summarized in the section "Three-Tiered Strategy."

Networking activities
The Reliability Subcommittee explored reliability best practices inside and outside the chemical sector and the company and reported to the network in December 1997. It recommended the use of 12 reliability practices and six reliability tools:
Reliability practices

• Elements of preventive maintenance
• Equipment repair history
• Corrosion monitoring
• Portable vibration monitoring
• On-line vibration monitoring
• Infrared thermography
• Positive material identification
• Rotating equipment alignment
• Steam trap monitoring
• Lube oil analysis
• Rotating equipment balancing
• Critical equipment monitoring

Reliability tools

• Reliability in engineering
• Reliability modeling
• Equipment maintenance plans
• Root cause failure analysis
• Reliability centered maintenance
• Data recording and analysis

All documents were placed on the Amoco Web page and updated as needed. The Reliability Subcommittee developed a self-assessment process for the plants that serves as the basis for "scorecards" developed by a new Measurements Steering Committee. The subcommittee's official task is complete, but the company continues to benefit from the relationships and networks that exist between the plant representatives, and the group plans to meet at least twice a year to share successes and problems they are experiencing in the reliability arena.

The Pumps Subcommittee also met extensively in 1997 and 1998 to explore and recommend best practices relating to pump maintenance and reliability in the same fashion as the Reliability Subcommittee. It completed its work in 1998 and issued seven best practice documents: pump repair procedures, pump repair documentation, pump repair training, condition monitoring, preventive maintenance, mechanical seals, and root cause failure analysis.

These documents are on the Amoco Web page, and the group also made recommendations to the Measurements Steering Committee to follow progress of the implementation of these practices. Amoco continues to benefit from the relationships and informal networks that remained in place.

The two other subcommittees produced best practice documents that were distributed throughout the chemical sector.

Benchmarking
The benchmarking process used by Edwin K. Jones, P.E., Inc. is conducted in three steps and focuses on a model of seven best practices:

• Leadership
• Planning & scheduling
• Preventive & predictive maintenance
• Reliability improvement
• Spare parts management
• Contract maintenance management
• Human resource development and training

The first stage of the assessment process was a kickoff meeting at each plant. It was designed to form the plant's benchmark team, define the roles of team members, review the benchmarking process, and provide an initial tour of the plant. A data questionnaire was left for the benchmark team to complete.

The second stage was a two-day meeting referred to as the validation visit. During this stage, key data are validated to be consistent with the comparison database. There are also interviews with maintenance craftsmen, maintenance supervision, operators, operator supervision, stores employees, reliability/maintenance engineers, contractor supervision, training coordinators, and maintenance planners and schedulers. A preliminary, verbal report of the findings is made to the Plant Leadership Team at the conclusion of the validation visit.

At this point, interpreting the comparison data, the interview issues, and the plant condition is initiated with discussion among team members. These issues are included in the final report, along with the observations of the "unbiased, external, calibrated resources," highlighting the opportunities for improvement. The final report has a balanced mix of team observations of maintenance practices and validated comparison data displayed in graphic plots along with data from "World-Class" plants.

The final stage, another two-day meeting, takes place approximately one month after the validation visit. The plant usually receives a benchmark report about a week before the third visit. The report includes plant data compared with other Amoco plants and with a selected set of "World-Class" plant data. Also provided is an initial estimate of the potential savings that might be obtained if the plant could close the gaps with the World-Class plants.

This third visit concludes the assessment process with a plant-wide review of the benchmark report. A great deal of emphasis is placed on shifting from the "assessment mode" to a "strategy development mode." The basic concept is: "OK, now that we have a better idea of where we are, where do we go from here?" The second day of this visit is then devoted to jump-starting the beginning of a Maintenance and Reliability Strategic Plan. The team selects areas to be improved, then lists specific tasks to deliver the desired results. Champions, resources, and dates are assigned to each task. The benchmark team then completes the strategic plan development over the ensuing two to three months.

Benchmarking results are summarized in the section "16 Plants Benchmarked." Each site's plan usually includes an overview of the maintenance strategy and how it fits with the plant's overall manufacturing strategy. It also includes some analysis of the savings potential that results from executing the plan. Savings are viewed in two categories: maintenance cost savings and business benefits of improved equipment reliability and availability. The plan lists the areas of improvement, the key issues, specific actions to be taken, metrics to be tracked, and a Gantt chart showing a timeline for all of the tasks to be completed. Seven plants have presented their plans to the network, with the remaining plants scheduled to do so in 1999.

Total Productive Maintenance (TPM)
After a TPM presentation to the Manufacturing Council in July 1998, the TPM Steering Committee (TPMSC) started to develop details around the essential TPM elements Amoco planned to pursue. These elements included:

• Equipment improvement teams (EIT)
• Asset ownership (autonomous maintenance)
• Asset reliability
• Maintenance effectiveness
• Early equipment management
• Training

The TPMSC included in its TPM model all known best practices from the subcommittees, the pockets-of-excellence from the benchmarking initiative, and other best practices gleaned from networking with other TPM companies and consultants. Because much of what was included already existed somewhere in Amoco, the TPMSC developed a site assessment tool to enable the plants to assess the quantity of work required in each of the TPM elements. This enabled plants to develop short-term strategies for the elements that were in use, and longer-term strategies for the elements requiring more time and effort.

The TPMSC then selected four areas of TPM as good places to start. They were autonomous maintenance/clean to inspect; process recording and data entry (PRIDE), an operator-based data gathering process using a combination bar code reader and data entry tool; equipment improvement teams; and work order prioritization.

These "places to start" provided a means to quickly immerse the plant in a TPM culture using best practices that had proven successful within Amoco. Plants could quickly link up with other plants that had implemented a given process, learn from their experience, and generate early success while developing their longer-range plans.

"Clean to Inspect," a process taught by Productivity, Inc., a TPM consulting firm, works on the principal that as you clean equipment, you inspect it thoroughly in the process. The detailed inspection identifies small abnormalities that could lead to poor performance or a breakdown. In other words, if you take care of all the little things, the big things will take care of themselves. It also has several ancillary benefits in that the employees who return equipment to like-new condition will work to keep it in like-new condition. During the process, cross-functional teams look for opportunities to improve the performance and maintainability of the equipment. It has been very successful at two sites in transforming operators from merely operators of the equipment into equipment caretakers.

PRIDE is an Amoco data-gathering tool that also can transform the operator into an equipment reliability resource. Using the Equipment Specific Maintenance Plans (ESMP), employees can select the equipment reliability data to monitor the equipment condition, predict impending failures, and help troubleshoot the root cause of breakdowns. This information can then be trended and used by the equipment improvement teams.

Equipment Improvement Teams (EIT) exist in some form at most of the plants. They are cross-functional teams that are given the time, training, and resources to address the root cause of poor equipment performance or breakdowns. Several plants have well-established EIT programs that serve as the cornerstone of their TPM effort. Other plants use them sporadically to solve major problems. The intent of TPMSC is to upgrade plants' use of EIT.

As cited in the benchmarking report, work order prioritization, misuse, and abuse were barriers to plants being able to plan and schedule their work. The Texas City plant recognized this in 1996 and learned a process called the Ranking Index of Maintenance Expenditures (RIME) at a planning and scheduling workshop at the Marshall Institute. It is a process where criticality of the equipment and the importance of the different types of work determine priority with minimum interference from people. Its use dramatically reduced the number of urgent work orders and it was cited as a pocket-of-excellence during benchmarking. It was thought that by using RIME, all plants could do more and better planning and scheduling to reduce costs and take a first step away from reactive unplanned maintenance.

The TPMSC also researched job postings for TPM coordinators, external TPM consultants, and Amoco technical experts for specific elements and tools, as well as suggested training and training material. All of this was assembled into a TPM Manual to assist the plants with developing their implementation plans.

TPM workshops
TPMSC realized there was a need for a wider base of TPM knowledge throughout the sector. As plants discussed TPM with their employees, inaccurate statements were causing resistance to the initiative. A three-day workshop was designed to teach the attendees the fundamentals of TPM, let them meet people who were already doing TPM, see TPM in action at a plant, learn about site implementations, and learn about "good places to start." It was hoped that the attendees could then return to their plants and accurately describe TPM and start to develop implementation plans.

By September 1998, five more workshops were held. In all a total of 325 people were trained, giving the sector the knowledge base it needed to move forward.

TPM implementation plans
Most plants now have TPM implementation plans. Several sites have appointed full time TPM coordinators, most have EIT, half are committed to installing PRIDE, and six have contracted with external resources and have done "Clean to Inspect" training. More than half has instituted RIME.

The TPMSC continues to meet every two to three weeks, usually via teleconference, to discuss progress. It advertises successes and assigned action items to investigate areas where progress is slow or lacking. A TPM activity scorecard is used to track progress and TPM metrics are being incorporated into computerized maintenance management system (CMMS) software. In 1999, the TPMSC plans to initiate a TPM Users Forum so people from across the sector can come together to share successes, work on common issues, and look for help in problem areas. The committee will make detailed assessments of the issues at each of these sites to develop the topics for the forum.

The TPMSC has asked MAN to further develop a vision and training for Asset Reliability/Reliability Engineering and for Maintenance Effectiveness/Planning and Scheduling. MAN commissioned these steering committees in August 1998. These visions are described in the section "Planning and Scheduling" and the section "Reliability Engineering."

Best practices implementation
MAN established the Best Practices Steering Committee (BPSC) to facilitate implementation of the best practices recommended by the Reliability and Pumps Subcommittees and to investigate new best practices for new issues. The BPSC charter ncludes:

• Place best practices in the "top drawer" of people who use them everyday. Make available expert maintenance resources to all levels of personnel. Provide a Web page search engine to obtain easy access to these practices and resources.
• Best practices and resources will be investigated and recommended to the plants by the BPSC. Each plant will be urged to evaluate the strategic fit of these practices and tools to their business. If applicable, they will perform a cost/benefit analysis to determine the priority of implementation.

Metrics and measurements
Because the plants used different computer systems and accounting practices, getting consistent unmanipulated data for measurements has been a major problem for MAN. With the impending start up a new company-wide system, MAN commissioned the Measurements Steering Committee (MSC) to develop standard measures. Its objectives are:

• Establish standard CMMS cost reports for all U.S. chemical plants.
• Consolidate MAN best practice scorecards for the purpose of measuring progress.
• Select and define key CMMS metrics to be used and recommend targets.
• Investigate reporting for non-CMMS and non-U.S. plants.

Effects of strategy
The most dramatic effect of MAN has been the downward trend in maintenance costs. Cost performance has closely followed the goals set forth in the three-tiered strategy and was reduced by 30 percent in two years. However, the improvements in reliability measures did not materialize as quickly as expected but their goals are still expected to be met.

Several of the activities sponsored by MAN should deliver additional benefits in 1999 and 2000. The Maintenance and Reliability Strategic Plans at each of the plants were completed in 1998 and should have a substantial impact as they are fully implemented. Most of the pump and reliability best practices have been migrated to the plants and are beginning to have an impact on performance. Plants will be implementing their TPM plans, and Reliability Engineering will be significantly upgraded in 1999.

Together these two initiatives will fundamentally change maintenance in Amoco's chemical sector from reactive to proactive and should dramatically impact the availability ratio and OAE for Amoco Chemical's manufacturing assets. MT

Edwin K. Jones is principal of Edwin K. Jones, P.E., Inc., 28 Quartz Mill Rd., Newark, DE 19711; (302) 234-3438; e-mail JJones1432@aol.com. Mark E. Lawrence is an internal maintenance and reliability consultant and is the coordinator for the Maintenance and Reliability Network (MRNet) at BPAmoco, WL4 Room 1794B, 200 Westlake Blvd., Houston, TX 77079-2682; (281) 560-4411; e-mail lawrenme@bp.com.

Manufacturing and production enterprises are under intense pressure to achieve maximum efficiency. The winners will be those that use their people and equipment assets most effectively. The objective is to optimize the utilization of all plant assets, from entire process lines to individual pressure vessels, piping, process machinery, and vital machine components.

Optimized asset profiles
But what does an optimized asset look like? To start on the road to optimization, it is vital to first decide on the metrics to define the state of optimized asset utilization. Input must be solicited from the engineering, operations, reliability, maintenance, purchasing, safety, regulatory, and risk management departments within an organization to develop these Optimized Asset Profiles (OAP). Members of this multi-disciplinary team bring their own slices of information to the group to use in specifying the required performance/uptime measurements and maximum cost metrics for each service location.

The OAP must be closely aligned with business objectives. The use of financial benchmark metrics is essential for gaining senior management support. A comprehensive OAP considers the total life-time financial impact that any asset installed at a service location has on production, quality, safety, hazardous waste disposal, and costs of maintenance, conversion, inventory, insurance, purchasing, installation, overhaul, and final disposal.

In the power generation industry, OAP metrics for a steam turbine generator system might include:

• Asset kilowatt hour (kWh) output per year
• Asset operations cost (steam cost, control systems, and labor) per kWh
• Asset maintenance cost (preventive maintenance, health monitoring, spare parts, labor, major overhaul, and inventory) per kWh
• Asset utilities cost (feed water and auxiliary electrical costs) per kWh
• Asset insurance cost per kWh
• Asset abnormal situation cost (annualized estimate of unbudgeted safety risks and mechanical failure risks) per kWh

One paper producer is attempting to measure daily asset profitability for each paper machine. The equation to calculate this number is:

Daily Asset Profitability ($) = Income from Sellable Tons Produced Today  All Daily Costs

The following items must be considered in the daily costs:

• Cost of capital today
• Cost of raw materials for tons produced today
• Operations cost today
• Maintenance cost today
• Utilities cost today
• Insurance cost today
• Unplanned event risk cost today

Information required for optimization
Because business requirements change daily, the OAP also will change daily and need continuous refinement. Evaluating the current state of each process equipment asset versus its OAP also requires a constant feed of information. Performance, reliability, and asset health analysis is regularly needed in order for operations and maintenance to make adjustments to assets in order to align with the OAP.

An example of the need for continuously updated OAP data is the new deregulated power generation marketplace where power stations need to track the hourly cost of each kilowatt-hour of electricity they generate and compare it to the current market price. As the price of each kWh drops, the profitability of operating a higher-cost plant diminishes. If plants have implemented OAP metrics, management has access to the information it needs to make timely, optimum decisions.

Information paths that have an influence on optimized equipment asset utilization are illustrated in the accompanying diagram. Each data node on the optimized asset utilization star has important asset information which needs to be synchronized with other data nodes and merged into comprehensive asset information. The goal is to make timely and informed decisions to safely and profitably maximize the value of the respective assets.

The data domains that affect asset optimization include the following sectors:

• Engineering design and configuration management
• Operations planning
• Safety, regulatory, and insurance compliance
• Process execution
• Reliability planning and analysis
• Maintenance execution
• Asset health monitoring and analysis
• Inventory, MRO purchasing, and financial

Engineering design and configuration management
Design specifications for the process equipment and its function in a plant process or machine train are fundamental to asset management. Process and instrumentation diagrams, drawings, manuals, and revision histories for plant production processes and equipment functionality are obviously required to maximize the use of equipment assets. The configuration management data provide the understanding of the design of the process and the specification for the purchasing of the proper asset, which will meet the tolerances of the system without wasting energy resources.

Quantitative process and component-level risk assessments are also an important guide to understanding the most likely failure modes and the effects of these failures on the plant. These assessments can then guide the condition monitoring and nondestructive testing efforts. If a piece of equipment fails and a new one needs to be ordered, the design specification of the failed component is required to be certain that the new replacement asset meets all operational requirements.

Safety, regulatory, and insurance compliance
Understanding the safety issues related to the installation, operations, and maintenance of the equipment assets and production processes is also crucial. Safety concerns affect decisions on when to perform certain high-risk repairs and how long to operate an asset which is in critical need of repair.

Although the environmental regulatory requirements which govern the process (government-required pollution controls, hazardous material disposal, local noise regulations, etc.) may not vary on a daily basis, their proper interpretation is vital in making decisions related to operations and maintenance. An asset optimization team must understand these limits as it makes decisions to run, reduce loading, shut down, or schedule various maintenance tasks. An understanding of equipment repair and inspection regulations, such as regulatory pressure vessel codes, jurisdictional inspections for boilers and pressure vessels, and insurance-driven inspections, also is necessary.

Decisions regarding equipment assets can affect the availability, cost, terms, and conditions of purchasing insurance. A plant with a high asset failure history will normally be underwritten differently than one with a low frequency of significant failures. Understanding the business aspects of a process is essential to properly operating and maintaining it to an optimum level.

Operations planning
The asset optimization team needs to know the current, planned, and historical production and efficiency levels of the process line assets and have the ability to plan modifications to these levels. Demands for the current time period must be balanced with future needs. In some cases, over-production carries a penalty because of warehouse limitations or other business factors.

The incoming process inputs, fuel, electricity, steam, cooling water, etc., need to be reviewed for availability and quality. This is especially important where the process input varies widely. Scheduled downtime or turnaround times are also important information.

Many companies are turning to enterprise resource planning (ERP) systems as the core information technology for their operations. These systems can tie all aspects of the supply chain together, usually within a single data warehouse structure. Production planning and scheduling is one of the important functions of the ERP system.

Process execution
The operations execution plan with the actual scheduling of manufacturing personnel and production equipment is an important source of data. Understanding the backlogs or bottlenecks in the current production stream can assist in asset optimization.

The process control and monitoring systems are commonly called distributed control systems (DCS). These production control systems work in conjunction with local programmable logic controllers to control a process and monitor its current state. The actual production data on current load, speed, temperature, and other process variables are essential to understanding the current health of an asset.

Quality assurance stipulations, including compliance with ISO 90xx standards, are required in many industries. Data regarding the quality of manufactured goods during the production process should be accessible to the asset optimization team.

Reliability planning and analysis
Reliability planning is a vital step toward improving asset utilization. The process begins with the identification of appropriate business metrics, which then are communicated to the asset optimization team for tracking. Typical targets for improvement include unscheduled production downtime and slowtime, maintenance overtime costs, and the cost of spare parts.

After setting the targets for improvement, the next step involves the study of the criticality of all production assets related to their impact on future production requirements, safety, regulatory compliance, spare parts costs, and unplanned failure costs. The definition of what constitutes a failure for a piece of equipment is normally broadened to include speed and load reductions that impact production. Structured approaches such as reliability-centered maintenance (RCM) facilitate this study. The results of this study involve a customized maintenance plan for equipment assets, specifying an optimized combination of condition based maintenance (predictive maintenance), time/usage-based maintenance (preventive maintenance), and failure-based maintenance (reactive maintenance). The output of this analysis will shape reliability-driven maintenance execution and reliability feedback activities.

Reliability analysts also regularly review failures and near misses. After a failure occurs, reliability analysts perform structured root cause analysis to determine the causes of the failure, and modifications are then made in the reliability plan to prevent future occurrences. This might include monitoring additional factors that could have signaled impending failure. An enterprise asset reliability system captures the reliability plan and logs the failures. The system facilitates reliability studies using a variety of software tools.

Maintenance execution
Maintenance execution processes should be based on the output of the preceding reliability planning step. An enterprise asset management (EAM) system or computerized maintenance management system (CMMS) assists in planning and scheduling maintenance manpower and tools. Traditionally, the maintenance organization was defined by the number of major overhauls it could staff and manage, the overtime hours worked against budget, wrench time, control of backlog, and emergency response. Now, the targets are not activity-based, but focused on reliability metrics, increased production output, and expense controls.

Information related to maintenance execution is of great interest to the asset optimization team. Much of the required data relates to the asset nameplate data (manufacturer, model, and specification), maintenance tool availability, and problem histories. Work order planning, scheduling, and tracking are other data sources. This information is useful for knowing what steps have been completed and then specifying the future direction of maintenance activities.

Asset health monitoring and analysis
Most machine and process characteristics which affect quality, availability, capacity, safety, risk, and cost can be continually evaluated throughout the life of an asset. Because this information is a vital feedback loop to modify the current reliability plan based on real-time signals, enhanced reliability organizations are now focusing attention on finding signs of impending failure.

The actual conditions of an asset are conveyed through various sensors. Asset health monitoring, also called condition monitoring, measures critical areas from the reliability plan which were designated as requiring condition based maintenance. Monitoring techniques include vibration signature analysis, lubricating oil analysis, electrical circuit analysis, and thermographic imaging.

The current health of process equipment forms another important node of information for the asset optimization team. Data from operations, protection, on-line condition monitoring, and off-line condition monitoring systems are needed in order to synchronize the various signals for diagnosis and prognosis of asset health.

Best results are obtained by collectively evaluating a complementary mix of characteristics, selected to provide the most accurate measure of overall condition on the specific type of equipment. Specialized analysis tools such as operating deflection shape analysis, virtual sensor analysis, and transient data analysis assist the equipment analyst.

Operators do not normally desire the detailed raw monitoring data gathered by the various technologies, but do require an integrated health analysis, augmented with clear recommendations and forecasts. New enterprise asset health (EAH) systems combine all available health monitoring data to assist an analyst in recognizing abnormal patterns and diagnosing problems. These enterprise-wide systems contain a large database of all condition monitoring indicators and sophisticated multi-parameter alarming techniques. They also provide a platform for automated analysis and communication of health advisories and action requests to a process control system, an enterprise asset maintenance system, and an enterprise asset reliability system.

Inventory, MRO purchasing, and financial
The inventory of replacement assets and spare parts currently on site or in storage is important to the asset optimization equation. The asset optimization team requires information from this area in order to optimize the specification of additional spare equipment and parts. An oversupply of spares takes up costly storage space and ties up excess capital. However, too few spares could cause a lengthy production downtime.

The maintenance, repair, and operations (MRO) purchasing department houses information on preferred vendor arrangements and lead-time-to-delivery of replacement parts. This information avoids rush purchases and allows just-in-time delivery of parts. Cost savings from the use of preferred vendors are also facilitated.

Financial systems tie inventory, purchasing, labor, and materials costs together for management reporting. These systems normally need to be fed information on the utilization of the asset and the manpower utilized.

Data integration issues
Integration of data between various asset software systems is the key to providing timely information to decision-makers to safely and profitably manage equipment assets. The challenge of communicating with the same language between engineering design (CAD and parts library), operations planning (ERP), operations execution (DCS), reliability, maintenance execution (EAM and CMMS), asset health monitoring and analysis (EAH systems), inventory, purchasing, and financial systems is formidable.

Most software providers within each system group are accustomed to a single information and functional structure. Many lack awareness of the potential value of information from other sources and characteristics that must be accommodated to gain full value.

Today, most systems store their information in a proprietary database format with little concern for external program requests for maintenance histories, spare parts availability, and failure events. Process control systems generate archive log data, each with its own file format and structure. Complementary condition measurements are typically gathered by separate systems that cannot communicate or share data for collective comparison. The process is so difficult, expensive, and time-consuming that vital comparisons to confirm accurate status and to predict lifetime are seldom made.

The results of vibration analysis, oil analysis, and other crucial tests are not easily available to a complete machinery diagnostic/prognostic expert system, or to a maintenance system for maintenance or operations to be adjusted. Data from the process control system are not readily available to a vibration condition monitoring system to analyze exception events. The needed link between engineering, enterprise resource planning, process control, maintenance management, and asset health systems has never materialized for many plants and is thwarting the promise of optimum asset utilization.

The value of integration The value of integrating asset systems can be seen from documented results at the largest power-producing utility in the United States, Southern Co. In a 2-year integrated monitoring and maintenance pilot system across five plants, the company has documented more than 100 instances in which information available from equipment health analysis was used to influence maintenance decisions.

During slightly over 1 year, potential savings and avoided costs of about $1 million resulted from deferring planned maintenance on healthy machines and from identifying problems in time to schedule repairs and avoid equipment failures.

In one case, time-based preventive maintenance was eliminated for major plant fans. Technicians now rely on vibration, oil, motor current, and temperature condition based monitoring techniques to determine which fans, gearboxes, and motors to maintain. In one planned plant outage, this resulted in savings of 340 man-hours because the fans did not have to be individually inspected for potential repairs. This reduced maintenance hours associated with these fans by 54 percent.

Open path to integration
Industrial users who desire to integrate their systems have three choices: attempt to buy all software from one vendor, launch in-house integration efforts, or utilize industry-standard open system interfaces from equipment software providers.

Arguably, there is not one software provider that provides a complete system for optimum asset management today. Some companies offer a much broader array of software with a single database platform that will operate with each other.

Today, users who want to provide interoperability between their plant information systems seem to be left with little choice except to hire a system integration company to piece the various systems together. This is usually cost-prohibitive and requires regular updates to the glue software any time one vendor's database format changes.

The cost of developing custom links between various systems can be extremely high and the cost of maintaining each of these links has been estimated at an annual rate of 40 percent of the initial development effort. For example, an $800,000 integration effort will require an on-going annual cost of $320,000 for software maintenance and upgrades.

A better long-term strategy is to purchase systems that utilize industry-standard open system interfaces. The benefits include:

• Lower cost for electronic exchange of vital information between proprietary systems
• Freedom to assemble plug-and-play information systems from multi-source best-for-application components
• Assurance of continuing least-cost upward growth and expansion to gain maximum advantages from improvements in knowledge, technology, practice, performance, and product features
• Increased value through maximum use of economical high performance consumer components with proven reliability, multi-source support, and rapid evolution to meet requirements of larger markets
• Reduced need for costly integration software
• Reduced integration software maintenance costs

Partners for integration
Currently, there are four organizations that are addressing industrial standards for system integration: International Standards Organization (ISO), Open Applications Group (OSG), OPC Foundation, and Machinery Information Management Open Systems Alliance (MIMOSA). In a system integration project, a company should review the specifications from these groups to see if an open protocol can be utilized. An overview is provided in the section Organizations for Open Systems Standards. The choice of suppliers who will partner with a plant in supporting its goals for optimizing the utilization of its assets is critical. Plants may wish to avoid software systems that utilize proprietary closed databases and architectures. A plant should consider the following questions before purchasing manufacturing technology systems:

• Does the supplier have a history of providing systems that are open and utilize industry standards wherever possible?
• Do the modules I purchase from this supplier integrate among themselves, possibly using different databases and architectures?
• Are the products I purchase certified as compliant with current industry standards?
• Are the software modules I am purchasing from this vendor costly to interface to other systems?

Plants are being forced to achieve higher profitability. To increase a plant's profitability, the optimized utilization of equipment assets is essential. To perform this optimization, plants require timely access to integrated data. The cost of this integration is normally a barrier to many plants moving to this optimization level. The use of open systems is lowering this barrier and allowing plants to begin to make strategic use of vital information. Manufacturing companies who commit to partnering with suppliers who provide systems that support open integration architectures will speed down the road to optimizing their valuable equipment assets. MT

Ken Bever, who serves on the board of directors of MIMOSA, is strategic project manager, Advanced Enterprise Systems Group, ENTEK IRD International, 1700 Edison Dr., Milford, OH 45150; telephone (513) 576-6151; e-mail kbever@entek.com; Internet www.entek.com

It is 1999. The maintenance function has come of age. Its performance can be planned, measured, tracked, and im-proved. The data it generates can be captured, analyzed, disseminated, and integrated. What's next? Will it be driven by corporate need or by information technology (IT) capabilities?

Where we are now and where we are headed are both defined by desired abilities: upgradability, integratability, scalability, and (stretching the metaphor a bit) mobility. Interestingly, it is easier and less confusing to describe the future than it is the present, which only points out the necessity of clarifying the technological state of the industry before leaping to what is on the horizon.

Let's start with the more easily solved problems. Forward-looking users of computerized maintenance management systems (CMMS) and enterprise asset management (EAM) systems want scalability, a system that can grow with them and accommodate hundreds of power users and thousands more non-maintenance, casual, or self-service users interfacing with the system. But this is only one side of the coin. The world of mergers and acquisitions is also a world of divestitures and spin-offs. While there has been an understandable focus on being able to move from a departmental to an enterprise system in order to get full value from it, users should keep in mind that it is also quite possible that a division of a well-integrated company could just as suddenly find itself an independent operation or the adopted child of a very different parent organization.

For these reasons, I believe scalability needs to imply movement toward both larger and smaller organizations. Clearly, there are products available now for any size organization, but the ideal, from both a cost and a flexibility perspective, is a system that can scale in both directions without changing code. More EAM/CMMS suppliers will need to develop this capability in response to what I predict will become a growing market need.

Besides providing the flexibility to operate as a stand-alone or highly integrated operation, single-platform systems offer another benefit that is just being recognized, particularly by smaller companies. That is the availability, at no extra cost, of special, sophisticated functionality originally developed for other companies, often at great expense. Using departmental asset maintenance software that springs from a single code base as an enterprise product opens up possibilities for expansion otherwise unavailable to departmental users. And having the latest functionality, like having the newest technology, is often an important competitive advantage.

The quest for mobility
Another frequently mentioned item on many maintenance wish lists is mobility, but the lack of clarity about just what mobility means contributes to the confusion that surrounds the capabilities available in today's market. Often mobile computing implies wireless technology, although this is not always the appropriate solution. The technology itself is less important than the benefits it bestows, namely untethered real-time data transmission.

Fully wireless, fully mobile (FWFM) solutions for field-dispatched workers, e.g., cellular digital packet data (CDPD), have become quite attractive recently with the introduction of fixed-price, virtually unlimited-usage communications contracts from network vendors and handheld device batteries that last an entire 8-hour shift. With only one exception, FWFM solutions are still on the drawing board in asset management because of the rigorous demands of lightweight application development.

Semi-wireless, real-time solutions such as radio-frequency and infrared LAN connections may be affordable in limited-duty applications such as storeroom management and central-plant equipment logging, though wiring a large, spread-out LAN with wireless nodes every 500 to 1000 ft can be too costly for a single application like maintenance. Operating at higher communication speeds though, such applications are correspondingly less demanding to develop and a number of asset maintenance software users are implementing such systems.

What is more widely and affordably available are store-and-forward solutions that capture data and transmit it in batch format at a later time. These are the heaviest of the mobile solutions because entire tables of reference data, e.g., spare-part availability and equipment failure history, must be carried by the mobile unit for reference and validation purposes, something of no concern to a system with real-time access. Such solutions are quite common today, but rarely use wireless connections because of the amount of data synchronization involved in their operation.

Mobility is understandably associated with the Internet. Most solutions utilize the Internet in some way, although the plain old telephone system is still heavily used. Even though Web-enabled sounds like a leading-edge buzzword, fast and cheap Internet access by EAM systems is today's reality. The major competitors are already Web-enabled. Solutions will be refined, products will become more widely available, and costs will come down, but mobility and remote access are here now.

Integration, upgrading, open systems
Integration and upgrading have been the rocks on which all enterprise business applications, including EAM, have foundered for years. Fortunately, the solution to both lies in standards-based, distributed component architecture. Note that I did not say simply component architecture. The difference is comparable to Mark Twain's comment: The difference between the right word and the almost right word is the difference between lightning and the lightning bug. It is important to be clear on this point, as all vendors may say they have components, each meaning something different by it. (In fact, the word component has been used in programming circles for years.) But it is in an architecture reliant on standards-based, distributed-component technology where the real promise, and real dollar benefits to customers, lie.

The first response to the problem of integration came with the development in the 1980s of applications from different vendors that could be made to work together. The process was difficult, time-consuming, and often hideously expensive, but it could be done. And it was such a significant development that it was given the moniker open systems architecture. However, open is a relative word; one can walk through an open door but not an open window.

Open systems were like windows; they allowed systems to shake hands, but they did not really allow easy entrance, and so integration has continued to hinder a corporation's ability to choose the specific applications it really needs. The decision often has come down to opting for the best functionality in the most critical systems, compromising in other areas. For many organizations, this has meant choosing a tier-one financial or back office system bundled with other functional modules that are second tier at best. For others it has meant enduring the time, expense, and irritation of integrating different applications.

Although the concept of open systems made integration possible, it did nothing for the problem of upgrading. Generally, there has been no easy way to add functional enhancements without waiting for an entire new release of the product, a time frame that can span several years. And new releases have always been a big bang proposition; there has been no way to add only selected functional enhancements. The problem has been further compounded by the fact that many prior customizations (especially database reconfiguration) must be reprogrammed for the new release. Again, it is industry leaders who have made the greatest effort to minimize both the difficulty and the expense of customization.

DCA and barrier-free systems
Standards-based, distributed-component architecture (DCA) can finally eliminate the windowsills and even most of the doorsills of integration. And it can facilitate the easy adoption of enhancements as they become available. In short, DCA ushers in a new era of barrier-free systems. But the components must be standards-based. They must share a common language that enables them to talk to each other.

As the confusion between components, distributed components, and standards-based components suggests, the development of DCA has been an evolutionary process. There are other terms that further confound the issue. For instance, do loosely coupled enterprise application suites differ from DCA? If the business logic uses standards-based (i.e., nonproprietary) messaging to communicate among modules, then no. But DCA does differ dramatically from tightly coupled business application suites which use proprietary (i.e., single-vendor) messaging. While such applications are designed to work easily with modules by the same vendor or through a vendor-supplied, proprietary interface, they are monolithic and somewhat inflexible when it comes to interfacing with the applications of other vendors. And when it comes to upgrading, the effort is a carefully orchestrated IT operation and is often accomplished only at enormous cost.

Bringing down TCO
One of the key measurements spoken of today in IT departments is total cost of ownership (TCO). Although the phrase is relatively new as a software term, the issues it raises are familiar ones: integrating and upgrading, and the cost of doing both. The use of standards-based, distributed-component technology will enable customers to upgrade on the fly, immediately as new functionality becomes available. Cost savings or efficiency improvements associated with that functionality need not be delayed months or years until a new release is issued by the vendor and scheduled by IT. There are also cost savings associated with being able to add only desired enhancements and in the ease of upgrading. The term associated with this capability is slipstreamed upgrades.

How important are slipstreamed upgrades? In today's era of heightened competition, can you afford to have your systems compromised or down for a massive upgrade? Can you afford to delay an upgrade and let your competitors bypass you in the critical area of information technology? Certainly Wal-Mart's erstwhile competitors paid the ultimate price for not having the real-time inventory and stocking systems in which Wal-Mart wisely, if at high initial cost, invested.

Standards-based DCA also means that the time, expense, problems, and frustration associated with integration eventually will become things of the past. All back-end integration today is either database dependent or proprietary; the integration software that works with Microsoft SQL Server must be completely rewritten for an Oracle database, and integration using messaging middleware relies on proprietary formats. With standards-based distributed components, integration will no longer be constrained in this regard, eliminating not only the financial cost but the competitive cost of non-optimal systems and the risk associated with having critical operations interrupted. Ask anyone who's been through it what the disruption caused by an enterprise systems integration project is like, especially one that fails to achieve its target return on investment.

What's taking so long?
In sum, standards-based DCA virtually eliminates the most costly and troublesome issues associated with enterprise business application software, including EAM systems. This provides users competitive advantage and better protection against business interruption. It is a heady promise, but it has just begun to be realized. No vendor offers a complete, distributed-component-based product & yet. But several tier-one enterprise software vendors are committed to DCA technology; these are the vendors that already offer one or more standards-based distributed components and have publicly announced their intention to convert completely to standards-based DCA in the future. If DCA is the answer to so many prayers, why hasn't everyone jumped on the bandwagon? The answer is not hard to find. As with any revolutionary technology that threatens to supplant a well-established one in which thousands have made huge investments, and with so-called network technologies in particular, it will take time for DCA and its benefits to acquire critical mass. Owning one of the first fax machines brought few benefits because there were few people to whom you could send your fax messages.

In the case of DCA, two things have stood in the way of quick acceptance: the lack of agreement on standards and the lack of available commercial development tools. Only in the past year have standards such as DCOM, CORBA, JDBC, EJB, and XML become mainstream business application development platforms and tools such as Visual J++, BEA Web Logic, and Visual Age become generally available. The windows of open systems are giving way to the doorways of barrier-free systems, but the true barrier-free era will begin when everyone, not just one vendor, gets rid of the windowsills.

Defining the leaders
Having conquered client/server and open systems architecture, enterprise integration, and functionality issues, vendors who would be leaders in asset maintenance now should be committing to standards-based DCA. Change may be the order of the day in many regards, but some things will remain constant. From the time only two decades ago when maintenance was just beginning to define its value to the corporation, customers have continued to seek vendors with certain gold-standard qualities. They want a vendor that will meet their needs today and one that has the financial strength and the vision to invest in the future. Look for companies that have those very qualities.

Certain companies take the lead in exploring lightweight application technologies such as DCA, in developing consistently robust functionality, and in making their product as flexible and as widely applicable as possible. While niche players focus well on meeting the current functional needs of their particular customer base, leaders are committed to the customer's long-term benefit, whether measured in dollars or performance. Think of an everyday example. All cars run well on their first day out. The best ones still run well 100,000 miles later & but require far less to operate and maintain along the way. MT

Milton Bevington is manager of product marketing for PSDI, 100 Crosby Dr., Bedford, MA 01730; (781) 280-2000; Internet www.maximo.com