6

Equipment Efficiency: Availability, Performance, and Maintenance

The role of maintenance is to ensure the survivability and proper functioning of all company hardware (productive and nonproductive). Most maintenance departments are considered, by most companies, “a necessary evil” or a money pit that represents a continuous cost. Managing a maintenance department at times can be nearly impossible because the investments required to improve production processes usually take on a low priority or, even worse, may not even make it to the priority list for capital expenditures.

Maintenance evolution, as well as maintenance techniques evolution, has been developed in parallel for many companies: The first obligation of the maintenance department is to remediate hardware failures that have occurred already. The next obligation after fixing breakdowns is to prevent future problems with the equipment that eventually may lead to failure.

The companies that are most advanced in maintenance management try to incorporate basic maintenance tasks into their daily production routine such that direct labor personnel check fluid levels and examine production equipment for potential failure mechanisms, also searching for ways to increase the ability to predict potential equipment breakdowns.

EQUIPMENT MAINTENANCE

The primary objective for the maintenance department of any company is to maintain the productive resources at a high operative level in order to ensure their service at an expected cost. There are several terms that attempt to explain why equipment maintenance is important. These terms are helpful, but no expression is as clear as the one that defines maintenance as the “machine's medicine” (Fig. 6.1).

According to this analogy, the maintenance department is in charge not only of fixing the machines (or correcting improper performance) but also of carrying out checkups concerning the machines' well-being in order to avoid breakdowns (prevent improper performance). The maintenance mission is to keep the equipment in good working order and also to determine the right moment to replace equipment. When the predicted replacement time is reached, the effort to keep the machine running is no longer cost-effective, and it is better simply to take it offline.

The elements that need maintenance in a factory are numerous. They include direct (machine tools, welders, etc.) and indirect production equipment (test equipment, coolant distribution equipment, etc.). Basically, all available equipment requires maintenance, even if it is not used directly to carry out productive tasks:

• Machines (mechanical, electrical, and pneumatic parts) and tools
• Facilities (compressed air, heating, electrical systems, etc.) and buildings (walls, illumination, etc.)
• Information and transportation systems (if they belong to the company)

Some companies subcontract maintenance, whereas other companies manage maintenance of their own resources. A good example of this situation is the fact that most companies now subcontract their facilities and building maintenance, e.g., mowing the lawns, general landscaping, and parking renovations. In cases, the maintenance service is fully subcontracted. In other cases, there is no on-board staff to deal with machine breakdowns.

Figure 6.1. Maintenance is the “machine's medicine.”

Efficient maintenance management and equipment conservation are contemplated in both just-in-time (JIT) and the 20 keys (key number 9) for lean manufacturing (Fig. 6.2). Breakdown reductions improve the availability and performance rates of the equipment (Fig. 6.3), thereby improving overall equipment efficiency.

THEORETICAL BASIS

Types of Maintenance

All industrial equipment is exposed to transitory (wear) or definitive breakdowns (catastrophic failure) affecting its functionality and performance. In many cases, equipment failures can represent high costs for enterprises, as well as high risks for the workforce; therefore, they must be solved or repaired as soon as possible.

However, the maintenance mission is not just to repair the breakdowns. Maintenance should be able to get ahead of the breakdowns. Both tasks correspond to the two major types of maintenance: corrective and preventive.

Preventive maintenance has two variants:

• Systematic preventive maintenance
• Conditional preventive maintenance or predictive maintenance

Figure 6.2. Location of maintenance in just-in-time and 20 keys diagrams.

Figure 6.3. Maintenance improves availability and performance rates.

Corrective Maintenance. Until the 1950s, corrective maintenance was virtually the only maintenance carried out in companies, also called breakdown maintenance. During those years, machine stoppages hardly affected productive time. The machines were less complex and more reliable, and the repairs, mainly in the mechanical parts, were carried out in an effective way.

Figure 6.4 illustrates the time evolution for the level of performance of a particular piece of equipment when applying corrective maintenance only. There are two types of corrective maintenance:

• Urgent repairs. This type of repair is for machines that have had a breakdown, causing the machine stop. In most cases, returning the equipment to service has a higher priority than fixing the machine properly. In other words, the repair that is carried out is provisional one, and the machine downtime should be as small as possible. The remaining tasks needed to repair the machine fully will be scheduled for a future time.

  

  Figure 6.4. Behavior of a component with corrective maintenance.

• Scheduled corrective. This type of maintenance appears as a result of urgent repairs. When machine service is reestablished, it is necessary to determine an appropriate time to repair the machine completely. After repairing the damaged component, it can be as good as new or at least as it was before (e.g., replacing a light bulb or repairing a flat tire).

The main problems with performing repair tasks are

• Repair tasks are performed quickly and under pressure, which can cause future problems.
• Repair time can be very high because replacement part(s) needed to get the machine running again may have to be ordered from a supplier.
• Accidents can take place because of poor maintenance safety measures.

Keeping only a corrective maintenance policy implies higher labor costs, especially when several machines are damaged at the same time. In this case, enterprises have to deal with high-idle-time issues. However, this policy can be justified in some cases, e.g., when maintenance is only performed on equipment with a frequent replacement policy (such as office computers) or when breakdown costs are small (light bulbs fail).

Systematic Preventive Maintenance. In the early 1960s in the United States, General Electric Corporation systematized a new type of maintenance called planned maintenance. This type of maintenance focused on improving certain machine aspects after breakdowns had been repaired. Four years later, planned maintenance arrived in Japan, where the bases of this systematic preventive maintenance process were established. At first, this maintenance consisted of only a systematic substitution of some machine components (that still were working properly) with new components.

This type of maintenance is applied to general wear or use components, such as bearings and filters, and is used on equipment with high failure costs. This systematic replacement policy is carried out to avoid breakdowns. Besides the systematic replacement, it is also very important to know with precision the component's performance characteristics (curve) in order for this type of maintenance to be effective (Fig. 6.5).

Component replacement can be carried out continuously. For example, a typical replacement policy might call for a component to be replaced every week (even though this method is getting obsolete) or in other ways, such as every 300 working hours or every 1000 parts produced.

Conditional Preventive Maintenance or Predictive Maintenance. Systematic preventive maintenance can become very expensive because several components can be replaced despite still being in good operationing condition. Conditional preventive maintenance is a method that is used to change components based on their current state (Fig. 6.6). Using this maintenance policy, the useful life for costly components can be extended.

For conditional or predictive maintenance, it is not critical to have very accurate component performance curves; however, this type of maintenance best fits components for which performance can be monitored (via product attributes) or judged by the operator. It is therefore necessary to have a strict and detailed method for analyzing process data, noticing the actions required for specific conditions. ISO 14,000 (corresponding to the environmental norm) requires that industry avoid systematic preventive maintenance when working with oils and environmentally harmful products and encourages conditional or predictive maintenance.

Figure 6.5. Behavior of a component with systematic preventive maintenance.

Figure 6.6. Behavior of a component with conditional preventive maintenance.

Sometimes, it is possible to monitor, in real time, a component's state and determine when working anomalies are about to occur. In this case, maintenance workers must proceed with immediate component replacement. Moreover, such maintenance generally is called predictive maintenance.

The principal difference between conditional preventive maintenance and predictive maintenance is that the operating variables are adjusted constantly using conditional predictive maintenance (large families of parts may be produced). It becomes necessary to look for correlations between multiple parameters and the degradation of a component.

Some variables that can be measured are temperature (thermocouples), noise (phonometer), cracks (x-ray machine), and pressure losses (manometer). The normal QS9000 recommends predictive maintenance procedures, and in the case of not having them, it is necessary to justify the economic problems of investing in them.

MAINTENANCE PROGRAM IMPLEMENTATION

Prior to selecting the best methodology to improve equipment availability (reducing the number of breakdowns), it is always a good policy to first find the root of the breakdowns, as well as the current state of the company (maintenance-wise). Almost all machines follow a similar life cycle from installation until disposal:

• Hidden small defects. It is difficult to detect the appearance of these defects because they are hard to see and are not apparent at first. This type of defect normally is unattended and most likely will go unfixed owing to the fact that it does not interfere with functionality. An example would be increased friction in an axle.
• Apparent small defects. These types of defects are more noticeable, e.g., small vibrations on a machine. When the defects do not affect the equipment's functionality directly, they are normally not repaired either.
• Execution under expectations. In this case, the defects affect equipment productivity. Most likely the machine will not work accurately, and the standards of quality will be violated. Often it is difficult to find the problem or cause because the specific damaged component is unknown.
• Intermittent stops. The machine intermittently produces defective parts, and there are numerous process settings. Small repairs that get the machine working are performed, but in a provisional manner.
• Stops and breakdowns. The machine's performance is poor, and breakdowns are frequent. Breakdowns can become expensive in terms of lost production time and money (Fig. 6.7).

In some companies, production equipment may not be as new as one would like it to be. This equipment may have already gone through some of the stages just described and most likely is at the “stops and breakdowns” stage. Replacement or renewal (rebuilding) should be considered here.

Another problem that firms face is that production equipment typically becomes more sophisticated, faster, and more expensive each year (i.e., longer payback or extended capitalization). The result of purchasing new equipment is that the equipment has greater economic impact (higher product volume), and therefore, repairs should be done at a faster rate. Working shifts also can represent an obstacle to maintenance interventions. The implementation of working shifts (up to five in some cases) limits possible maintenance tasks and scheduling.

Figure 6.7. Poorly maintained equipment.

The objective of maintenance is to efficiently oversee equipment throughout the equipment life cycle through all the stages described in this section. In order to be successful, it is necessary to cover the entire life cycle, beginning with implementing an effective corrective maintenance plan, then preventive maintenance tasks, and later on, predictive maintenance strategies.

Getting Started

The first step, before any maintenance policy implementation, is to become familiar with the resources that will require maintenance. Each maintained resource should be coded (resource ID) in order to distinguish it from the rest of the resources because it is not unusual to have identical pieces of equipment.

It is also important to code the types of breakdowns, as well as the maintenance tasks performed. As a result, in a historical data study, it is possible to group failure causes of each piece of equipment in order to take proper corrective actions. There is a specialized taxonomy that helps to code resources and breakdowns, although most of the time common sense and simplicity lead to a good coding system.

After this preliminary phase, maintenance should have the following two documents:

• Facility inventory. This document lists all the plant equipment and its principal characteristics: code, record number, equipment type, etc.
• Equipment history files. This is a paper (or electronic data) portfolio that contains the data given by the equipment manufacturer, information about location of the equipment in the plant, an outline or a picture of the equipment, the characteristics of the equipment installation, and finally, the types of spare parts needed.

Corrective Maintenance Implementation

For corrective maintenance, the number one priority is to organize, in an effective way, the corrective maintenance procedures and actions. When a breakdown occurs, the worker who has discovered it should fill out a breakdown work order. This document will be different in each facility because the information needed in each maintenance department varies.

If the worker can solve the problem, he or she must fill out a report to the maintenance department, so that the breakdown gets registered; otherwise, the work order will be sent to the maintenance department, and a work request order will be issued.

Depending on the work urgency, maintenance workers either will repair the machine immediately or will schedule the repair. The repair can be provisional or definitive. In the first case, it is recommended to keep the work order open until the repair is completed.

Scheduled Corrective Maintenance. Variability in the duration of corrective maintenance tasks can be problematic if they are not scheduled properly. In order to complete repairs efficiently, corrective orders and flow diagrams for repetitive repairs must be developed explaining how the repair should be done and listing the materials and spare parts that should be used. Adequate maintenance time should be allocated for all repairs.

When maintenance workers carry out corrective operations, the task does not end with the equipment repair. There are some other procedures that must be followed. Maintenance workers should gather all the breakdown information (Fig. 6.8) and describe the process that was performed to solve the problem and document them in the work order.

Each machine should have its own file with breakdown records; otherwise, it can be very difficult to analyze breakdown causes and anticipate future problems accurately. This file must be upgraded with each maintenance intervention; therefore, the work of maintenance workers does not end when the resource is running again but when they have fully registered the breakdown in the records. For frequent airline travelers, this action is the all too familiar process of fixing the switch or the signal light in the cabin (5 minutes) and then waiting while all procedures and issues are signed off and cleared (20 minutes). In cases where life is at risk (as in air travel), repair, report, acknowledge, and then recheck constitute common practice. Corrective maintenance tasks consist not only of changing the broken or malfunctioning components but also of studying the causes and frequency of the breakdowns. This is the most appropriate time to introduce into a company the concept of preventive maintenance.

Figure 6.8. Maintenance workers should gather all the breakdown information.

Preventive Maintenance Implementation

Of all of the numerous goals of preventive maintenance, the number one concern is to avoid a breakdown of any resource while keeping maintenance cost as low as possible. Preventive maintenance includes two types of actions:

• Inspections. The objective of this action is to observe and detect possible anomalies. Generally, inspections consist of frequent checkups, sometimes daily, that follow a specific inspection plan.
• Revisions. These actions require scheduled equipment stops and usually suppose a systematic substitution of several machine components.

In companies with machines that run in sequence, scheduled stops usually are carried out over the weekend (or during nonproductive periods), increasing maintenance costs. Preventive maintenance task scheduling is mandatory in the ISO norms. These tasks can be planned daily, weekly, monthly, or even annually. They always should be scheduled at times when they do not affect the factory's production plan. However, the reality is that daily working problems force us to continuously reschedule these tasks. It is still worthwhile to schedule these preventive maintenance tasks and to try to adhere to this schedule.

Equipment preventive maintenance tasks are also called PM orders. Each PM order should be based on a study of the causes of equipment breakdown. This study can be carried out by using the equipment FMEA tool described in the tools section later in this chapter.

Autonomous Maintenance

Maintenance staff normally will carry out most of the PM orders; however, some PM orders can be carried out by production workers as part of batch setup or shift startup. This group of orders is known as user maintenance orders.

These user maintenance orders are the key in the development of autonomous maintenance, and they should be simple and graphically represented, indicating to the workers the exact place of all the elements that should be inspected and the tasks that should be carried out (Fig. 6.9).

Again, the calendar for autonomous maintenance repair tasks (selfemployed maintenance) is basic because many inspection tasks should be carried out every day. Therefore, the maintenance department may receive a significant number of notices that could be handled easily by production worker (e.g., a dirty optical switch, a plug that does not work, etc.). In some cases, it takes more time to fill out the request order than to fix the problem. Autonomous maintenance includes these small tasks and three daily preventive measures: cleaning, lubricating, and checking (Fig. 6.10).

Security for autonomous task execution is one of the most important restrictions when determining the type of activities that production workers can perform safely (Fig. 6.11). Repair or maintenance never should be performed by the operator if the knowledge required to fix the machine is high. For complex maintenance, a repair specialist should take care of problem. Autonomous tasks apply only for simple repair operations. The maintenance department staff should perform complex maintenance tasks as well as dangerous tasks.

Sometimes it can be very challenging to convince production workers about the importance of maintenance tasks. Companies face this issue because workers do not want to perform tasks that they do not consider their responsibility. This situation is similar to checking the oil in a car. Most of the time, the oil level is okay, and the check is not necessary. Even worse, the maintenance department may not allow operators to perform maintenance tasks on the equipment.

Figure 6.9. User maintenance orders must be easy for production workers to understand and accomplish.

Figure 6.10. Cleaning, lubricating, and checking the equipment.

The autonomous maintenance process has a specific methodology (Fig. 6.12) made up of seven steps that lead to its complete implementation.

TPM: Total Productive Maintenance

In the 1970s, Nakajima developed total productive maintenance (TPM) in Japan. TPM was a new maintenance management philosophy included in Toyota's improvement process.

Although the TPM concept was introduced in Japan in 1971, it took a long time to be published in English. In fact, the first English translation was not published until 1988. TPM development resulted in the creation of the Japanese Institute of Plant Maintenance (JIPM), which grants a PM prize to top companies in TPM. Sixty percent of the winning companies over the first 17 years are now part of the Toyota group or suppliers of this group.

Nakajima combined preventive maintenance theories with the total quality concept. As a result, Nakajima developed the overall equipment efficiency (OEE) ratio (studied in Chap. 1), which is one of the TPM keys:

Figure 6.11. Security is one of the main restrictions in autonomous task assignment.

Figure 6.12. Autonomous maintenance implementation methodology.

• Maximize the overall equipment efficiency. This is done by eliminating the six big losses described in Chap. 1 (breakdowns, setup and changeover, idling minor stoppages, reduced speed, defects and rework, and starting losses).
• Autonomous maintenance implementation. This is done in order to terminate the “I operate, you repair” mind-set.
• Preventive engineering. This is done to avoid the need to carry out maintenance operations on equipment, improving its maintainability (see reliability concepts in the tools section of this chapter).
• Training workers for maintenance improvements. Workers who operate the machines should be able to propose methods for increasing equipment availability by eliminating breakdowns or expediting repairs.
• Initial equipment management. This is done in order to avoid negative effects from the machine setup process.

Nakajima proposes specific objectives in each category of equipment losses and develops, mainly, the activities related directly to machine maintenance. All the PM prize-winning companies have an OEE ratio greater than 88 percent.

Although setup time and initial run quality, to mention but a few targets, affect a machine's efficiency, they are addressed in detail in another specific methodology that will be explained in a later chapter. In the maintenance field, the objective of TPM, or the ideal situation, is zero breakdowns. To achieve this goal, it is necessary to use such tools as P-M analysis, which is presented in the maintenance tools section later in this chapter.

RCM: Reliability-Centered Maintenance

RCM was created in the United States in the 1960s to optimize the reliability of aeronautical equipment. RCM was not used in nuclear power stations until the 1980s (after the Three Mile Island accident) and has been implemented only recently in the rest of the industrial world.

In order to begin RCM implementation, it is necessary to have a complete maintenance and breakdown record for each piece of equipment. Starting from this equipment information, the RCM objective is to determine maintenance tasks that are most effective for critical components. There is a specific methodology that facilitates RCM's implementation in companies based on several tools, such as FMEA, reliability analysis, statistical techniques, etc.

To apply RCM efficiently, it is necessary for a company to have a preventive maintenance program that runs properly. The starting point for RCM is performing a statistical analysis of equipment behavior owing to breakdowns; thus, without enough statistical information to characterize breakdowns, RCM cannot be applied adequately.

MAINTENANCE TOOLS

FMEA for Equipment

All defects have a root cause, and to eliminate future defects, an action must be carried out. For example, if the defect is in the gap between two elements (contact between them instead of clearance), then the cause or root of the problem can be lack of lubrication or a loose fastener. The action in each case would be different (i.e., grease or tighten the lose element).

To establish a good preventive maintenance plan, all the possible breakdowns, their causes, and their corrective actions must be analyzed. The main tool to carry out this type of analysis is failure mode and effects analysis (FMEA) for equipment.

The FMEA tool (Fig. 6.13) is a guide to analyzing, in an organized manner, causes of possible equipment breakdowns. In order to avoid breakdowns, a group of workers is gathered to study the problems and failures that can take place on the equipment. This team establishes action plans to avoid the failure causes discovered.

In an FMEA study, the team should enter data for each of the fields corresponding to each column in the template (see Fig. 6.13):

Figure 6.13. FMEA template.

• Equipment functions. In this column, all the functions that the equipment carries out are registered. For example, (1) manufacture specific parts, (2) provides compressed air during specific conditions, etc.
• Failure modes. Through brainstorming, all the possible ways that the equipment can be forced to stop (breakdowns) are determined. Each stop is related to one of the functions previously registered, e.g., breaks, blockage, leaks, etc.
• Failure effects. All possible consequences of each failure are analyzed in detail. The most important factors to analyze are the effects and failure severity S for each failure. The impact or consequences that this failure can have is measured using a scale of 1 to 4, with 1 being not very serious and 4 being very serious. It is also indicated if the failure is critical or not for the company.
• Failure causes. The origin of the failure is analyzed. The goal consists on identifying the anomaly that can lead to the failure (e.g., low oil, a defective component, etc). It is important to estimate the probability P of each stoppage occurring (again on a 1 to 4 scale, with 1 being not very frequent and 4 being very frequent).
• Actual controls. This column indicates if, at the present time, some kind of control is carried out to avoid each cause. It also should indicate if the defect can be predicted and controlled (not the cause of the failure), that is to say, detection D (also on a 1 to 4 scale, with 1 being if the control does not always detect the cause and 4 being if it always detects).

After the first part of the FMEA table has been completed, the risk priority number (RPN) is calculated as the product of the three quantified variables (S, P, and D).

Using a Pareto diagram and ranking failures by RPN, the work team will analyze the causes that do not represent any threat and then choose those that could harm the process and try to eliminate them. Special attention must be paid to those effects that are considered critical, even though they may not have a high RPN.

The chosen plan of action and the employee responsible for carrying out this plan are registered in the same table used in the FMEA. Generally, after a FMEA application arises, the need for developing a preventive maintenance plan become apparent (Fig. 6.14).

In order to determine preventive maintenance intervention periods T, it is necessary to know the component-damage/wear-behavior curve. Information about breakdowns and the time when the breakdowns occurred is collected, and this facilitates the equipment reliability calculation in terms of mean time between failures (MTBF), as explained in the next section.

Figure 6.14. Preventive maintenance plan determination.

The first step is to calculate T based on the corrective percentage K that the company would like to support. Therefore, when corrective maintenance is being implemented, all the information related to the behavior of the components must be collected in order to predict future behaviors more accurately. It is also important to keep in mind that a premature or late preventive maintenance policy can be worse than corrective maintenance.

Reliability

Independent of the type of preventive maintenance plan implemented (systematic or predictive), the objectives of a maintenance plan are the same: Avoid equipment failures by increasing the reliability and effective life of the equipment at the lowest possible cost. Reliability is defined as the probability that a piece of equipment will work satisfactorily over a certain period of time under some specific working conditions. The following paragraphs explain these terms in depth.

Reliability is a probability. That is to say, it is not a deterministic measure of a component's useful life. This probability can be defined in many ways, although the most frequent way to define it is based on the relative frequency of breakdowns.

All production equipment should work satisfactorily. In other words, production equipment should not fail. Failure can be catastrophic or progressive. This means that failure can be triggered by an abrupt change in the component characteristic or by progressive damage. Since it is impossible to design and operate a production system that never has any breakdowns, the system should be designed and maintained to work satisfactorily for a specific period of time. This property arises because of the need for equipment elements to maintain quality standards over a reasonable period of time, since this duration cannot be infinite.

Reliability is a temporary variable, whereas quality is considered to be a momentary variable. Quality is considered momentary in the sense that it measures a product's characteristics and compares them with the product's specifications. However, reliability is the product's ability to maintain those specifications or characteristics throughout its useful life. Therefore, reliability can be defined as

The component or equipment life duration depends on working conditions. These conditions can be environmental (e.g., temperature or humidity) or operational (e.g., continuous starts and stops, electrical strain). Various components will not have the same reliability value under different working conditions.

The system state depends on the primary group of elements that makes it work properly, where each element has a random lifetime. Therefore, it is necessary to estimate the lifetimes of the components that wear out at a faster rate and to propose solutions so that component failures do not affect the entire system or equipment.

A good way to quantify the reliability is through the mean time between failures (MTBF) value:

Another important variable is maintainability, which is defined as the probability that if a breakdown has taken place, it must be repaired in a predetermined time following a specific repair procedure. Maintainability depends on different factors, such as

• Machine factors, such as accessibility or interchangeability among components
• Organizational factors, such as maintenance staff knowledge, documentation availability, or maintenance tasks subcontracting
• Operative factors, such as the ability of the staff and the clarity of working instructions

The maintainability is quantified through the mean time to recovery (MTTR):

Lastly, the statistical availability is an average between the middle time used in the equipment and the required production time:

If the different times between breakdowns and each repair duration time are represented graphically, it is easy to understand the statistical availability concept (Fig. 6.15).

When calculating this availability, setup times and other types of idle times should not be included. Because of this interpretation, the statistical-availability concept differs from the availability concept presented in the first chapters included in the OEE definition.

Bathtub Curved. Component manufacturers usually include, among their characteristics, the failure versus time performance of their products. It is not unusual for this plot to look like a bathtub curve (Fig. 6.16).

This curve is a graphic representation of the failure rate λ(t) that is directly related to product reliability. It can be defined as the probability that an element fails depending on its life use stage or status. The curve can be divided in three areas.

Figure 6.15. Graphic representation of time between failures and time to repair.

Figure 6.16. Bathtub curve example.

Zone I: Infant period. This zone coincides with the equipment setup and debugging process. It generally goes downhill because, as time moves forward, the probability of a component failure decreases. Problems in this area can be avoided by making intensive tests or by exchanging troublesome elements during an early-stage adjustment period. It is a vital stage, and maintenance workers have the responsibility of trying to shorten this period as much as they can while not affecting the component's useful life.

Zone II: Useful period. In this part of the curve, failures appear randomly. In electronic systems, there is no material wear; therefore, the curve is virtually horizontal. In mechanical systems, the curve normally has a slightly positive slope.

Zone III: Waste period. As a component or equipment reaches the end of its useful life, failures come far more quickly. In this stage, critical component replacement is strongly recommended.

P-M Analysis

The defects that take place on equipment result from two main causes: sporadic losses and chronic losses. Sporadic losses can be corrected using several tools. Some of these tools have been studied already, and others will be studied in following chapters. These losses are due to causes that can be analyzed and eliminated.

P-M analysis is responsible for eliminating chronic losses in equipment availability (Fig. 6.17), that is to say, those that are considered “natural” according to the root sources that drive them. For this reason, it is denominated P-M analysis (P = phenomenon; M = mechanism). For example, scratches on a part produced by close contact with another part with higher hardness are analyzed via P-M analysis.

In P-M analysis, the reliability that has been studied has two aspects:

Figure 6.17. Sporadic and chronic losses.

• Intrinsic reliability—due to the design and production of the component
• Operative reliability—due to component use and the maintenance process performed on it

P-M analysis should be applied after conventional improvement and has its own implementation methodology. We will not discuss this method any further in this book.

Recently, a new tool has appeared based on statistical practices, denoted as Six Sigma. Six Sigma is a systematic process that has become very popular (primarily because of application results). Six Sigma is also suitable to carry out this type of study, i.e., instead of doing a P-M analysis.

Maintenance Management

A maintenance department, like any other department, should manage and control its costs properly and make sure that planned activities have been carried out. Maintenance management also should plan for future objectives for extended periods.

There are an unlimited number of indicators that can be used for maintenance department performance. Each company will decide which indicators are more suitable for it: personnel performance, hours dedicated to urgent work, repair cost, availability, etc.

In many companies, maintenance management is a difficult task because it frequently does not have upper management's support. As long as the maintenance department does not exceed its assigned budget, no one pays much attention to maintenance department activities or expenses.

Maintenance Costs. Using economic terms, maintenance management helps to control deviations in a firm's budget as well as to determine investment needs to reduce costs. Most enterprises look for ways to keep their maintenance costs low.

There are two opposing alternatives in achieving this goal (Fig. 6.18): support the cost of carrying out maintenance tasks or support the cost of not carrying out those tasks:

• Nonmaintenance costs. These costs derive mainly from equipment breakdowns and equipment wear-out: opportunity costs, quality costs, production personnel costs, etc.
• Maintenance costs. Breakdown prevention costs, anomalies detection cost, inspection resources costs, etc.

In the first part of the curve (1), maintenance investment increases the equipment availability and at the same time reduces the nonmaintenance costs. In the second part of the curve (2), an increase in availability supposes large investments. The lower point on the total-cost curve corresponds to the optimal availability point.

Figure 6.18. Maintenance costs.

SUMMARY

This chapter has presented an overview of maintenance, a critical aspect of lean manufacturing. Maintenance planning and activities are determining factors for lean enterprises efficiency. Unfortunately, maintenance normally is perceived as a necessary evil and is not always seen as an engineering activity. This chapter has outlined some of the maintenance policies and procedures that can be used to obtain the goal of any production system: operating as efficiently as possible at the lowest cost.

RECOMMENDED READINGS

John Dixon, Uptime: Strategies for Excellence in Maintenance Management. Cambridge, MA: Productivity Press, 1995.

Salih O. Duffuaa, A. Raouf, and John Dixon, Planning and Control of Maintenance Systems: Modeling and Analysis. New York: Wiley, 1998.

JIPM: Autonomous Maintenance for Operators. Portland, OR: Productivity Press, 1997.

Kunio Shirose, Yoshifumi Kimura, and Mitsugu Kaneda, P-M Analysis: An Advanced Step in TPM Implementation. Portland, OR: Productivity Press, 1995.

François Monchy, Teoría y práctica del mantenimiento industrial. Barcelona: Asson, 1990.

Francisco Rey, Hacia la Excelencia en Mantenimiento. Madrid: TGP-Hoshin, S.L., 1996.

Seiichi Nakajima, Introduction to TPM: Total Productive Maintenance. Cambridge, MA: Productivity Press, 1988.