An introduction to predictive maintenance - part 2 pot

and finance experts. Financial evaluation of preventive maintenance is divided gen- erally into either single transactions or multiple transactions. If payment or cost reduc- tions are multiple, they may be either uniform or varied. Uniform series are the easiest to calculate. Nonuniform transactions are treated as single events that are then summed together. Tables 2–1 through 2–5 are done in periods and interest rates that are most applica- ble to maintenance and service managers. The small interest rates will normally be applicable to monthly events, such as 1 percent per month for 24 months. The larger interest rates are useful for annual calculations. The factors are shown only to three decimal places because the data available for calculation are rarely even that accurate. The intent is to provide practical, applicable factors that avoid overkill. If factors that are more detailed, or different periods or interest rates, are needed, they can be found in most economics and finance texts or automatically calculated by the macros in com- puterized spreadsheets. The future value factors (Tables 2–1 and 2–3) are larger than 1, as are present values for a stream of future payments (Table 2–4). On the other hand, present value of a single future payment (Table 2–2) and capital recovery (Table 2–5 after the first year) result in factors of less than 1.000. The money involved to give the answer multiplies the table factor. Many programmable calculators can also work out these formulas. If, for example, interest rates are 15 percent per year and the total amount is to be repaid at the end of three years, refer to Table 2–1 on future 38 An Introduction to Predictive Maintenance Table 2–5 Capital Recovery, Uniform Series with Present Value $1 Interest Periods 1% 2% 4% 10% 15% 20% 1 1.010 1.020 1.040 1.100 1.150 1.200 2 .508 .515 .530 .576 .615 .654 3 .340 .347 .360 .402 .438 .475 4 .256 .263 .275 .315 .350 .386 5 .206 .212 .225 .264 .298 .334 6 .173 .179 .191 .230 .264 .301 7 .149 .155 .167 .205 .240 .277 8 .131 .137 .149 .187 .223 .261 9 .117 .122 .135 .174 .210 .248 10 .106 .111 .123 .163 .199 .239 11 .096 .102 .114 .154 .191 .231 12 .089 .095 .107 .147 .184 .225 18 .061 .067 .079 .120 .163 .208 24 .047 .053 .066 .111 .155 .203 36 .0033 .038 .051 .094 .151 .200 48 .026 .032 .045 .092 .150 .200 60 .022 .028 .043 .091 .150 .200 CP P ii i n n = + () + () - Ê Ë Á ˆ ¯ ˜ 1 11 value. Find the factor 1.521 at the intersection of three years and 15 percent. If our example cost is $35,000, it is multiplied by the factor to give: $35,000 ¥ 1.521 = $53,235 due at the end of the term Present values from Table 2–2 are useful to determine how much we can afford to pay now to recover, say, $44,000 in expense reductions over the next two years. If the interest rates are expected to be lower than 15 percent, then: $44,000 ¥ 0.75% = $33,264 Note that a dollar today is worth more than a dollar received in the future. The annuity tables are for uniform streams of either payments or recovery. Table 2–3 is used to determine the value of a uniform series of payments. If we start to save now for a future project that will start in three years, and save $800 per month through reduc- tion of one person, and the cost of money is 1 percent per month, then $34,462 should be in your bank account at the end of 36 months. $800 ¥ 43.077 = $34,462 The factor 43.077 came from 36 periods at 1 percent. The first month’s $800 earns interest for 36 months. The second month’s savings earns for 35 months, and so on. The use of factors is much easier than using single-payment tables and adding the amount for $800 earning interest for 36 periods ($1,114.80), plus $800 for 35 periods ($1,134.07), and continuing for 34, 33, and so on, through one. If I sign a purchase order for new equipment to be rented at $500 per month over five years at 1 percent per month, then: $500 ¥ 44.955 = $22,478 Note that five years is 60 months in the period column of Table 2–4. Capital recov- ery Table 2–5 gives the factors for uniform payments, such as mortgages or loans that repay both principal and interest. To repay $75,000 at 15 percent annual interest over five years, the annual payments would be: $75,000 ¥ 0.298 = $22,350 Note that over the five years, total payments will equal $111,750 (5 ¥ $22,350), which includes the principal $75,000 plus interest of $36,750. Also note that a large differ- ence is made by whether payments are due in advance or in arrears. A maintenance service manager should understand enough about these factors to do rough calculations and then get help from financial experts for fine-tuning. Even more important than the techniques used is the confidence in the assumptions. Control and finance personnel should be educated in your activities so they will know what items are sensitive and how accurate (or best judgment) the inputs are, and will be able to support your operations. Financial Implications and Cost Justification 39 Trading Preventive for Corrective and Downtime Figure 2–7 illustrates the relationships between preventive maintenance, corrective maintenance, and lost production revenues. The vertical scale is dollars. The hori- zontal scale is the percentage of total maintenance devoted to preventive maintenance. The percentage of preventive maintenance ranges from zero (no PMs) at the lower left intersection to nearly 100 percent preventive at the far right. Note that the curve does not go to 100 percent preventive maintenance because experience shows there will always be some failures that require corrective maintenance. Naturally, the more of any kind of maintenance that is done, the more it will cost to do those activities. The trade-off, however, is that doing more preventive maintenance should reduce both corrective maintenance and downtime costs. Note that the downtime cost in this illus- tration is greater than either preventive or corrective maintenance. Nuclear power- generating stations and many production lines have downtime costs exceeding $10,000 per hour. At that rate, the downtime cost far exceeds any amount of mainte- nance, labor, or even materials that we can apply to the job. The most important effort is to get the equipment back up without much concern for overtime or expense budget. Normally, as more preventive tasks are done, there will be fewer breakdowns and therefore lower corrective maintenance and downtime costs. The challenge is to find the optimum balance point. 40 An Introduction to Predictive Maintenance Figure 2–7 The relationship between cost and amount of preventive maintenance. As shown in Figure 2–7, it is better to operate in a satisfactory region than to try for a precise optimum point. Graphically, every point on the total-cost curve represents the sum of the preventive costs plus corrective maintenance costs plus lost revenues costs. If you presently do no preventive maintenance tasks at all, then each dollar of effort for preventive tasks will probably gain savings of at least $10 in reduced corrective maintenance costs and increased revenues. As the curve shows, increasing the invest- ment in preventive maintenance will produce increasingly smaller returns as the breakeven point is approached. The total-cost curve bottoms out, and total costs begin to increase again beyond the breakeven point. You may wish to experiment by going past the minimum-cost point some distance toward more preventive tasks. Even though costs are gradually increasing, subjective measures, including reduced confu- sion, safety, and better management control, that do not show easily in the cost cal- culations are still being gained with the increased preventive maintenance. How do you track these costs? Figure 2–8 shows a simple record-keeping spreadsheet that helps keep data on a month-by-month basis. Financial Implications and Cost Justification 41 Figure 2–8 Preventive maintenance, condition monitoring, and lost revenue cost, $000. It should be obvious that you must keep cost data for all maintenance efforts in order to evaluate financially the cost and benefits of preventive versus corrective mainte- nance and revenues. A computerized maintenance information system is best, but data can be maintained by hand for smaller organizations. One should not expect imme- diate results and should anticipate some initial variation. This delay could be caused by the momentum and resistance to change that is inherent in every electromechani- cal system, by delays in implementation through training and getting the word out to all personnel, by some personnel who continue to do things the old way, by statisti- cal variations within any equipment and facility, and by data accuracy. If you operate electromechanical equipment and presently do not have a preventive maintenance program, you are well advised to invest at least half of your maintenance budget for the next three months in preventive maintenance tasks. You are probably thinking: “How do I put money into preventive and still do the corrective mainte- nance?” The answer is that you can’t spend the same money twice. At some point, you have to stand back and decide to invest in preventive maintenance that will stop the large number of failures and redirect attention toward doing the job right once. This will probably cost more money initially as the investment is made. Like any other investment, the return is expected to be much greater than the initial cost. One other point: it is useless to develop a good inspection and preventive task sched- ule if you don’t have the people to carry out that maintenance when required. Careful attention should be paid to the Mean Time to Preventive Maintenance (MTPM). Many people are familiar with Mean Time to Repair (MTTR), which is also the Mean Cor- rective Time (M — ct). It is interesting that the term MTPM is not found in any text- books the author has seen, or even in the author’s own previous writings, although the term M — pt is in use. It is easier simply to use Mean Corrective Time (M — ct) and Mean Preventive Time (M — pt). PM Time/Number of preventive maintenance events calculates M — pt. That equation may be expressed in words as the sum of all preventive maintenance time divided by the number of preventive activities done during that time. If, for example, five oil changes and lube jobs on earthmovers took 1.5, 1, 1.5, 2, and 1.5 hours, the total is 7.5 hours, which divided by the five events equals an average of 1.5 hours each. A few main points, however, should be emphasized here: 1. Mean Time Between Maintenance (MTBM) includes preventive and cor- rective maintenance tasks. 2. Mean Maintenance Time is the weighted average of preventive and cor- rective tasks and any other maintenance actions, including modifications and performance improvements. 3. Inherent Availability (A i ) considers only failure and M — ct. Achieved avail- ability (A a ) adds in PM, although in a perfect support environment. Oper- ational Availability (A 0 ) includes all actions in a realistic environment. 42 An Introduction to Predictive Maintenance Too many maintenance functions continue to pride themselves on how fast they can react to a catastrophic failure or production interruption rather than on their ability to prevent these interruptions. Although few production engineers will admit their continued adherence to this breakdown mentality, most plants continue to operate in this mode. 3.1 MAINTENANCE MISSION Contrary to popular opinion, the role of maintenance is not to “fix” breakdown in record time; rather, it is to prevent all losses that are caused by equipment or system- related problems. The mission of the maintenance department in a world-class orga- nization is to achieve and sustain the following: • Optimum availability • Optimum operating conditions • Maximum utilization of maintenance resources • Optimum equipment life • Minimum spares inventory • Ability to react quickly 3.1.1 Optimum Availability The production capacity of a plant is partly determined by the availability of produc- tion systems and their auxiliary equipment. The primary function of the maintenance organization is to ensure that all machinery, equipment, and systems within the plant are always online and in good operating condition. 3 ROLE OF MAINTENANCE ORGANIZATION 43 3.1.2 Optimum Operating Condition Availability of critical process machinery is not enough to ensure acceptable plant per- formance levels. The maintenance organization must maintain all direct and indirect manufacturing machinery, equipment, and systems so that they will continuously be in optimum operating condition. Minor problems, no matter how slight, can result in poor product quality, reduced production speeds, or other factors that limit overall plant performance. 3.1.3 Maximum Utilization of Maintenance Resources The maintenance organization controls a substantial part of the total operating budget in most plants. In addition to an appreciable percentage of the total-plant labor budget, the maintenance manager often controls the spare parts inventory, authorizes the use of outside contract labor, and requisitions millions of dollars in repair parts or replace- ment equipment. Therefore, one goal of the maintenance organization should be effec- tive use of these resources. 3.1.4 Optimum Equipment Life One way to reduce maintenance cost is to extend the useful life of plant equipment. The maintenance organization should implement programs that will increase the useful life of all plant assets. 3.1.5 Minimum Spares Inventory Reductions in spares inventory should be a major objective of the maintenance orga- nization; however, the reduction cannot impair their ability to meet the first four goals. With the predictive maintenance technologies that are available today, maintenance can anticipate the need for specific equipment or parts far enough in advance to pur- chase them on an as-needed basis. 3.1.6 Ability to React Quickly All catastrophic failures cannot be avoided; therefore, the maintenance organization must be able to react quickly to the unexpected failure. 3.2 E VALUATION OF THE MAINTENANCE ORGANIZATION One means to quantify the maintenance philosophy in your plant is to analyze the maintenance tasks that have occurred over the past two to three years. Attention should be given to the indices that define management philosophy. One of the best indices of management attitude and the effectiveness of the mainte- nance function is the number of production interruptions caused by maintenance- related problems. If production delays represent more than 30 percent of total 44 An Introduction to Predictive Maintenance production hours, reactive or breakdown response is the dominant management phi- losophy. To be competitive in today’s market, delays caused by maintenance-related problems should represent less than 1 percent of the total production hours. Another indicator of management effectiveness is the amount of maintenance over- time required to maintain the plant. In a breakdown maintenance environment, over- time costs are a major, negative cost. If your maintenance department’s overtime represents more than 10 percent of the total labor budget, you definitely qualify as a breakdown operation. Some overtime is, and always will be, required. Special pro- jects and the 1 percent of delays caused by machine failures will force some expen- diture of overtime premiums, but these abnormal costs should be a small percentage of the total labor costs. Labor usage is another key to management effectiveness. Evaluate the percentage of maintenance labor, compared to total available labor hours that are expended on the actual repairs and maintenance prevention tasks. In reactive maintenance manage- ment, the percentage will be less than 50 percent. A well-managed maintenance orga- nization should maintain consistent labor usage above 90 percent. In other words, at least 90 percent of the available maintenance labor hours should be effectively used to improve the reliability of critical plant systems, not spent waiting for something to break. 3.2.1 Three Types of Maintenance There are three main types of maintenance and three major divisions of preventive maintenance, as illustrated in Figure 3–1: • Maintenance improvement • Corrective maintenance • Preventive maintenance • Reactive • Condition monitoring • Scheduled Maintenance Improvement Picture these divisions as the five fingers on your hand. Maintenance improvement efforts to reduce or eliminate the need for maintenance are like the thumb, the first and most valuable digit. We are often so involved in maintaining that we forget to plan and eliminate the need at its source. Reliability engineering efforts should empha- size elimination of failures that require maintenance. This is an opportunity to pre-act instead of react. For example, many equipment failures occur at inboard bearings that are located in dark, dirty, inaccessible locations. The oiler does not lubricate inaccessible bearings as often as those that are easy to reach. This is a natural tendency, but the need for Role of Maintenance Organization 45 lubrication could be reduced by using permanently lubricated, long-life bearings. If that is not practical, at least an automatic oiler could be installed. A major selling point of new automobiles is the elimination of ignition points that require replacement and adjustment, introduction of self-adjusting brake shoes and clutches, and extension of oil-change intervals. Corrective Maintenance The little finger in our analogy to a human hand represents corrective maintenance (i.e., emergency, repair, remedial, unscheduled). At present, most maintenance is cor- rective. Repairs will always be needed. Better maintenance improvement and pre- ventive maintenance, however, can reduce the need for emergency corrections. A shaft that is obviously broken into pieces is relatively easy to maintain because little human decision is involved. Troubleshooting and diagnostic fault detection and isolation are major time consumers in maintenance. When the problem is obvious, it can usually be corrected easily. Intermittent failures and hidden defects are more time- consuming, but with diagnostics, the causes can be isolated and then corrected. From a preventive maintenance perspective, the problems and causes that result in failures provide the targets for elimination by viable preventive maintenance. The challenge is to detect incipient problems before they lead to total failures and to correct the defects at the lowest possible cost. That leads us to the middle three fingers—the branches of preventive maintenance. Preventive Maintenance As the name implies, preventive maintenance tasks are intended to prevent unsched- uled downtime and premature equipment damage that would result in corrective or 46 An Introduction to Predictive Maintenance MAINTENANCE IMPROVEMENT (MI) PREVENTIVE (PM) CORRECTIVE (CM) Reliability-driven Modification Retrofit Redesign Change order Equipment-driven Self-scheduled Machine-cued Control limits When deficient As requred Statistical analysis Trends Vibration monitoring Tribology Thermography Ultrasonics Other NDT Periodic Fixed intervals Hard time limits Specific time Breakdowns Emergency Remedial Repairs Rebuilds Predictive Time-driven Event-driven Figure 3–1 Structure of maintenance. repair activities. This maintenance management approach is predominantly a time- driven schedule or recurring tasks, such as lubrication and adjustments that are designed to maintain acceptable levels of reliability and availability. Reactive. Reactive maintenance is done when equipment needs it. Inspection using human senses or instrumentation is necessary, with thresholds established to indicate when potential problems start. Human decisions are required to establish those standards in advance so that inspection or automatic detection can determine when the threshold limit has been exceeded. Obviously, a relatively slow deterioration before failure is detectable by condition monitoring, whereas rapid, catastrophic modes of failure may not be detected. Great advances in electronics and sensor tech- nology are being made. Also needed is a change in human thought process. Inspection and monitoring should disassemble equipment only when a problem is detected. The following are general rules for on-condition maintenance: 1. Inspect critical components. 2. Regard safety as paramount. 3. Repair defects. 4. If it works, don’t fix it. Condition Monitoring. Statistics and probability theory are the basis for condition- monitoring maintenance. Trend detection through data analysis often rewards the analyst with insight into the causes of failure and preventive actions that will help avoid future failures. For example, stadium lights burn out within a narrow period. If 10 percent of the lights have burned out, it may be accurately assumed that the rest will fail soon and should, most effectively, be replaced as a group rather than individually. Scheduled. Scheduled, fixed-interval preventive maintenance tasks should generally be used only if failures that cannot be detected in advance can be reduced, or if dictated by production requirements. The distinction should be drawn between fixed-interval maintenance and fixed-interval inspection that may detect a threshold condition and initiate condition-monitoring tasks. Examples of fixed-interval tasks include 3,000-mile oil changes and 48,000-mile spark plug changes on a car, whether it needs the changes or not. This may be wasteful because all equipment and their operating environments are not alike. What is right for one situation may not be right for another. The five-finger approach to maintenance emphasizes elimination and reduction of maintenance needs wherever possible, inspection and detection of pending failures before they happen, repair of defects, monitoring of performance conditions and failure causes, and accessing the equipment on a fixed-interval basis only if no better means exist. Role of Maintenance Organization 47 [...]...48 An Introduction to Predictive Maintenance Advantages and Disadvantages Overall, preventive maintenance has many advantages It is beneficial, however, to overview the advantages and disadvantages so that the positive may be increased and the negative reduced Note that in most cases the advantages and disadvantages vary with the type of preventive maintenance tasks and techniques used... reduced Typically, predictive maintenance is implemented for one of the following reasons: • As a maintenance management tool • As a plant optimization tool • As a reliability improvement tool 4.1.1 As a Maintenance Management Tool Traditionally, predictive maintenance is used solely as a maintenance management tool In most cases, this use is limited to preventing unscheduled downtime and/or catastrophic... is important, predictive maintenance can provide substantially more benefits by expanding the scope or mission of the program As a maintenance management tool, predictive maintenance can and should be used as a maintenance optimization tool The program’s focus should be on eliminating unnecessary downtime, both scheduled and unscheduled; eliminating unnecessary preventive and corrective maintenance tasks;... predictive maintenance program Predictive Maintenance Costs The average maintenance budget of the plants interviewed was $ 12, 053,000, but included those with budgets ranging from less than $100,000 to more than $100 million The average plant invests 15.8 percent of its annual maintenance budget in predictive maintenance programs, but one-third (33%) of the plants interviewed in our May 20 00 survey... oncondition or condition-monitoring techniques is usually better than fixed intervals Advantages There are distinct advantages to preventive maintenance management The predominant advantages include the following: • Management control Unlike repair maintenance, which must react to failures, preventive maintenance can be planned This means “pre-active” instead of “reactive” management Workloads may be... Maintenance A side benefit of predictive maintenance is the automatic ability to monitor the meantime-between-failures (MTBF) These data provide the means to determine the most cost-effective time to replace machinery rather than continue to absorb high maintenance costs The MTBF of plant equipment is reduced each time a major repair or rebuild occurs Predictive maintenance will automatically display the... that is often required to correct improper or incomplete repairs Data acquired as part of a predictive maintenance program can be used to schedule and plan plant outages Many industries attempt to correct major problems or schedule preventive maintenance rebuilds during annual maintenance outages Predictive data can provide the information required to plan the specific repairs and other activities during... the maintenance staff to plan each repair The ability to predetermine the specific repair parts, tools, and labor skills required provided the dramatic reduction in both repair time and costs The ability to predict machine-train and equipment failures and the specific failure mode provided the means to reduce spare parts inventories by more than 30 percent Rather than carry repair parts in inventory,... either or both of the traditional maintenance programs (i.e., run -to- failure and preventive) Predictive maintenance can, however, reduce the number of unexpected failures and provide a more reliable scheduling tool for routine preventive maintenance tasks The premise of predictive maintenance is that regular monitoring of the actual mechanical condition of machine-trains and operating efficiency of process... work order and completion form Figure 3–4 Simple call report $ TOTAL $ 12. 75 $ 12. 75 CURR METER READ: 54 An Introduction to Predictive Maintenance COMPARISON (Benchmarking) SHORT-TERM TACTICS CURRENT (Maintenance Evaluation) GOALS VARIANCE (Gap Analysis) (Where you want to be and When) PROCESS & IMPLEMENTATION (How we get there) MEASURE (How we are doing) LONG-TERM STRATEGIES IDEAL (Duty-Task Analysis) . .179 .191 .23 0 .26 4 .301 7 .149 .155 .167 .20 5 .24 0 .27 7 8 .131 .137 .149 .187 .22 3 .26 1 9 .117 . 122 .135 .174 .21 0 .24 8 10 .106 .111 . 123 .163 .199 .23 9 11 .096 .1 02 .114 .154 .191 .23 1 12 .089. $1 Interest Periods 1% 2% 4% 10% 15% 20 % 1 1.010 1. 020 1.040 1.100 1.150 1 .20 0 2 .508 .515 .530 .576 .615 .654 3 .340 .347 .360 .4 02 .438 .475 4 .25 6 .26 3 .27 5 .315 .350 .386 5 .20 6 .21 2 .22 5 .26 4 .29 8 .334 6. .107 .147 .184 .22 5 18 .061 .067 .079 . 120 .163 .20 8 24 .047 .053 .066 .111 .155 .20 3 36 .0033 .038 .051 .094 .151 .20 0 48 . 026 .0 32 .045 .0 92 .150 .20 0 60 . 022 . 028 .043 .091 .150 .20 0 CP P ii i n n = + () + () - Ê Ë Á ˆ ¯ ˜ 1 11 value.