Minimizing Facilities Maintenance and Repair Costs Through Structured Management Techniques And Advanced Decision Support And Diagnostic Tools
Maintenance and repair (M&R) spending is a normal and expected cost of facility ownership. However, the cost can be minimized through an aggressive facility management program and the use of applicable diagnostic tools. The key to success is for managers to become proactive and embark on a realistic, long-range M&R program. Timely, well-planned M&R is cost effective. Lack of an effective M&R program results in reactionary (crisis) management as breakdown repair becomes the normal business activity. A reactionary approach results in high penalty costs that take ever-increasing amounts of the M&R and operating budgets. Consequently, more M&R is deferred, which in turn leads to accelerated deterioration and facility system and component failure.
FACILITY MANAGEMENT PROGRAM
A facility management program exists in every public works organization. In simple terms it is nothing more than the decision-making process. However, the decision-making methodology chosen will, in large part, determine success or failure. Ad hoc approaches tend to be reactionary and often lead to M&R programs that are costly on a per-unit basis. On the other hand, a structured facility management process can result in lower per-unit costs because the decision-making process is guided
by certain principles. These are discussed in this chapter, together with the two distinct management-level phases: network/facility and project.
Network/facility-level management encompasses many broad-based decisions. These decisions generally apply to entire networks or portions of networks (e.g., pavements, railroad tracks, pipelines) or facilities (e.g., buildings, piers, underground storage tanks). Network/facility-level management is performed on an ongoing basis and focuses on answering what, where, when, and budgeting questions. There are several aspects associated with management at this level. These include establishing programs for preventive maintenance, programmed maintenance, and planned (long-and short-range) specific M&R, all based on realistic budget projections. These are discussed in more detail later in this chapter.
Project-level management addresses the how best aspects of facility management. The purpose is to address the causes of distress associated with a given project, formulate feasible M &R alternatives, and choose the best one (or ones) for the project. Determining and correcting the cause instead of simply treating symptoms can reduce the rate of deterioration and stretch the time interval between corrective repairs.
Other key project-level management activities are the identification of all feasible M&R alternatives and the selection of the best one. Again, the nature and cost of the project will determine the level of effort associated with these activities. An economic analysis will help decide the lowest lifecycle cost alternative, although factors other than cost will sometimes govern the selection.
Project-level management is associated with short-range work plans. Only rarely will project-level management be accomplished for long-range projects. Continued deterioration and many other factors may negate the effort expended on project-level details if performed too early in the total management cycle.
COMPUTERIZED MAINTENANCE MANAGEMENT SYSTEMS (CMMS)
Effective facilities maintenance management requires the use and control of large amounts of information. This includes tracking incoming requests for facilities maintenance work, scheduling preventive maintenance, preparing budget estimates, projecting facilities maintenance requirements, and determining resource allocations. The input, storage, retrieval, manipulation, and display of this information is best accomplished by computers.
The past two decades saw the application of computer technology to facilities maintenance management expand as systems became more powerful, less costly, and easier to use. Automation products and processes evolved from pure record keeping and processing of accounting data to task automation and decision support. Sharing information interactively with other systems and direct system access by end users are realities.
Many commercial software products are available for use in facilities maintenance management. These systems are available for the full spectrum of computers to networked and stand-alone microcomputer systems. However, it is outside the scope of this committee's work to suggest specific hardware or software systems for use.
Preventive maintenance is the cornerstone of a solid cost-effective M&R program. Nominally, preventive maintenance is accomplished on a fixed schedule or frequency and consists of many check-point activities on items, most typically equipment. Examples include filter replacement, lubrication, and mechanical adjustments. Preventive maintenance reduces the risk of system or component failure, which if it occurs, would interfere with essential operations, endanger life or property, involve a high cost, or require a long lead time for replacement.
Not all facility systems or components should be included in a preventive maintenance program. These programs can be costly, so a preventive maintenance program should be reserved for those critical systems or components where the risk of failure must be reduced or where a preventive maintenance program can be shown to be cost effective on a life-cycle basis.
Systems or components not included in a preventive maintenance program are candidates for “run-to-failure” repair (unplanned), programmed maintenance (planned), or planned M&R based on condition and need. Typically, components included in a “run-to-failure” M&R strategy are small noncritical components that can be repaired or replaced on a service call. Other planned M&R strategies are described next.
PREDICTIVE TESTING AND INSPECTION (PT&I)
Predictive testing can reduce facilities maintenance costs and improve availability by enabling just-in-time maintenance of facilities systems and related equipment. Predictive testing monitors the condition or operating parameters of facilities system components to detect trends or conditions that indicate excessive wear or impending failure. This permits initiation of timely maintenance actions. There are embedded or on-line systems for continuous readings and portable systems for periodic readings. Software is available to integrate or interface predictive testing results with CMMS.
Some examples of predictive testing methods and applications are:
Oil analysis for wear metals and lubricant properties
Vibration analysis of bearings
Ultrasonic sound analysis for leaks
Infrared thermography scans of electrical equipment for heat buildup or loss and for roof and insulation failure
Motor circuit analysis for voltage imbalance
Magnetic signature or X-ray testing for ferrous metal failure
In addition, the use of PT&I during the commissioning of new or renovated buildings can detect defective equipment installation and defective equipment that otherwise appears satisfactory.
Predictive testing technology is evolving rapidly. A method to determine when to use predictive testing is to apply the Reliability Centered Maintenance approach described below.
RELIABILITY CENTERED MAINTENANCE (RCM)
RCM is a maintenance philosophy that incorporates the most logical and cost-effective mix of breakdown maintenance, preventive maintenance, predictive testing and inspection, and proactive maintenance.
Traditionally, many organizations practiced a mix of breakdown maintenance and preventive maintenance; some practiced breakdown maintenance entirely. Breakdown maintenance (also known as “run to failure”) rapidly consumed the life of equipment, which necessitated expensive replacements, and shut down operations without warning. Preventive maintenance was an improvement over breakdown maintenance and it extended the life of equipment. However, it was wasteful of resources, as the process was driven by time or operating cycles without regard to actual equipment condition. In addition, in times of constrained resources, preventive maintenance was the first area to be cut back or stopped due to the lack of any immediate significant impact or readily apparent damage.
In the 1950s and 60s the aircraft industry championed “on condition” maintenance, which utilized both simple and sophisticated nondestructive and noninvasive testing that could determine the actual condition and need for maintenance of a wide range of equipment. This new method evolved into what is now generally referred to as predictive testing and inspection. PT&I allowed an organization to reduce the amount of unnecessary maintenance performed, identify impending failure before it became catastrophic in nature, and allowed the scheduling of repairs and/or maintenance at a convenient time rather than on an emergency basis.
Proactive maintenance combines the principles of total quality management in order to identify the “root cause” of failures, such as defective parts, defective installation, improper operation, or unsuitable design. After the root cause is determined, actions are taken to remove or change the circumstance that caused the failure in order to gain the full expected life of other similar components.
Overall, reasonable application of RCM and its underlying processes will result in attainment of the full expected equipment life, reduced maintenance costs, and reliable operation of equipment with no significant adverse impact on safety or mission accomplishment.
Programmed maintenance is a planned M&R strategy that is viable when the proper recurrence frequency can be determined with a high degree of confidence, and it is cost effective on a life-cycle basis. Generally, the recurrence frequency is less often than that associated with preventive maintenance. Also, the risk associated with facility, systems, and/or component failure is less critical. Examples include pavement seal coats, roof drainage system cleaning from leaf accumulation, and certain painting.
PLANNED SPECIFIC MAINTENANCE AND REPAIR
Planned specific M&R consists of two phases: long range and short range. Long-range work planning is a network/facility-level management function and spans the entire multiyear planning horizon (typically 5 years). A long-range M&R plan should be updated periodically, at least annually. It is developed, in large part, from a condition survey inspection and condition assessment program, a prediction of future M&R needs, and a prioritization of total requirements.
Short-range work plans cover the present up to 1 year (typically called an annual work plan). The short-range plan is that portion of the overall plan that has “come due” and generally encompasses project-level management activities.
CONDITION SURVEY INSPECTIONS AND CONDITION ASSESSMENTS
Many facilities, systems, and components do not lend themselves to preventive maintenance, programmed maintenance, or a strategy of “run-to-failure.” These facilities, systems, and components should receive periodic condition survey inspections. The primary purpose of a condition survey is to collect just enough information to ascertain condition and provide a reasonable estimate of deficiencies. Long-range work plans are based on this information. The condition surveys need not be very detailed or performed very often. Sampling may be feasible for certain facilities. Anything more will increase the cost of inspection but only provide limited additional value. This is because survey information becomes “stale” due to the continuing deterioration process. Visual inspection techniques are
most often employed, and they may be supplemented with hand tools, meters, and sensors. Condition survey frequency is a function of facility type and use; system, component, and material type; current condition; and rate of deterioration. Depending on those variables, reinspection should occur every 1 to 5 years.
The condition survey inspection information must be translated into an assessment of condition. Generally, this involves the use of a condition rating. The rating should be objective, repeatable, sensitive enough to reflect condition changes, and correlate to budget requirements. Objectivity and repeatability are critical criteria for ensuring consistent ratings. Also, sensitivity is needed to accurately measure deterioration rates. The correlation to budgets will help to establish funding needs. An example of a condition rating procedure is the American Society of Testing and Materials (ASTM) standard for pavement evaluation: the Pavement Condition Index employs a 0 to 100 index scale. Different index values relate to specific M&R requirements and costs.
Fiscal realities require that M&R budgets be prepared many years ahead of actual M&R accomplishment. However, condition survey inspection and condition assessment programs only result in a listing of current requirements. Although some inspection findings will be deferred for future accomplishment, an accurate prediction of total future M&R needs is necessary to produce meaningful budgets. Condition prediction allows managers to find out what, where, and when facilities, systems, and components will need M&R. Models exist or are under development for predicting performance (or failure) or a future condition rating (or index) based on certain key variables.
Research has shown that as condition degrades the M&R cost for restoring the condition to an acceptable level increases curvilinearly. This implies that there is a “best” time to schedule M&R that results in the best condition at the lowest possible cost. Deviating from that “best” time results in a “penalty cost.” Robust predictive models can determine and list candidate facilities, systems, or components by year based on the predicted “best” time for M&R. Using cost versus condition correlations, an estimate of how much money is needed each year can be made.
PRIORITIZING MAINTENANCE AND REPAIR
Ideally, multiyear budgets could be prepared and funded based on the sum of the preventive maintenance, programmed maintenance, and specific long- and short-range M&R needs. A contingency amount could be added to account for “run-to-failure” and emergencies. Realistically, there are two major fallacies to this approach. One is that such a program would result in wide budget variations from year to year. The second is that the limited availability of public funds will not allow full funding of M&R programs every year.
Budget constraints mandate that the various elements of an M&R program be prioritized in a methodical manner. In the committee 's experience, ad hoc “squeaky wheel” approaches often result in poor allocation of resources. M&R program elements, including specific projects, can be ranked each year, based on cost (including penalty cost), mission needs, safety, environmental concerns, mandates, and/or any other important factors. Weighted guidelines models exist for setting these priorities. Operations research modeling tools, such as linear and dynamic programming, can also be used to optimally select a mix of M&R needs based on a set of constraints. Projects not selected for accomplishment in a given year are carried over into the next year, where they compete with new M&R requirements.
Often, the inspection process (condition survey or detailed inspection) relies heavily on the experience and observation skills of the inspector. However, many facility deficiencies defy visual detection until they become severe, if then. Fortunately, diagnostic tools exist to aid in both the condition survey (network/facility-level management) and detailed (project—level management) inspections.
Network/facility-level management condition surveys are used primarily to develop long-range work plans. Since these surveys do not need to be detailed, certain diagnostic tools can provide meaningful data quickly and cost effectively. Examples include infrared scans to detect wet insulation, pipe crawler cameras to detect buried pipe deterioration, and automated railroad track geometry measurements. On the “low-tech” side, the use of pen-based notebook computers has improved the gathering and
transfer of visually obtained data. On the “high-tech” side, video/photographic imaging and an array of built-in sensors are being developed to collect data in “real time” for automatic processing.
Sometimes, determining the cause of distress (project level) requires specialized diagnostic testing. Generally, determining the cause will involve an enhanced level of inspection detail not accomplished as part of the routine condition survey. The nature and cost of a project need to be weighed against the cost of the diagnosis and level of risk associated with no diagnosis or a misdiagnosis. At a minimum, except for the smallest of projects, an inspection is needed to clarify scope and establish project quantities. Many tools used for condition surveys are also useful for project-level diagnostics. Often, however, the project level-diagnostic tools need to be more sophisticated to help determine the cause and quantities of distress. Laboratory testing and load-carrying capacity analysis through nondestructive deflection testing are examples. In addition, results from ongoing, well-planned preventive maintenance and predictive testing and inspection programs can be utilized to a high degree in a condition assessment process or update, particularly if the data are contained in a state-of-the-art computerized maintenance management system.
Concurrent engineering is a concept to track, identify, investigate, and design out or purchase alternative products/equipment that is a better “fit” for a designated application. Misapplication or poor design can cause repeated and expensive failures that adversely affect the M&R program. New facilities must be designed not only for intended use but also for a low-cost maintenance program. Existing facilities need to be “reengineered” at the time of system or component repair/replacement or during rehabilitation to ensure that low-cost maintenance alternatives are considered. Close coordination with system/equipment operators and maintenance professionals is an important part of this process.