Free CMRP Exam Flashcards

All annual maintenance spend — maintenance labor (including operator maintenance hours), materials, contractors and services, plus outage, shutdown and turnaround work — divided by RAV and multiplied by 100. End-of-life replacement of worn-out machinery counts; capital spending for expansions or improvements does not, so the ratio compares upkeep rather than growth between plants of different size.

The cost to replace the plant's present production capability with new assets — process equipment plus the utilities, facilities and related assets that maintenance looks after — excluding land. It is not book, depreciated or insured value, so it does not shrink with accounting write-downs and stays comparable between sites.

Leading indicators measure the behavior that drives future results: PM and PdM compliance, percent planned work, proactive work share. Lagging indicators report outcomes already produced: downtime, MTBF trend, OEE and maintenance cost. You steer with leading indicators and confirm with lagging ones.

Everything from cradle to grave: purchase and installation, energy and operating cost, maintenance labor and spares, the production lost to downtime, and disposal or decommissioning. LCC is what justifies paying more up front for a more reliable design.

NPV discounts each future cash flow back to today's dollars at the company's cost of capital, so a multi-year savings stream can be compared fairly against the up-front spend. A positive NPV means the project clears the hurdle rate.

Payback (years) = initial investment ÷ annual net savings, and it stops counting at the break-even point. It ignores the time value of money and every dollar earned after payback, so a long-lived reliability project can look worse than a short-lived one; pair it with NPV or internal rate of return.

Because the repair invoice is only the visible slice. Each failure also costs lost production margin, scrap and quality losses, expediting and overtime, and safety or environmental exposure. Adding those up is what turns a reliability project from a maintenance expense into a business case.

So that two plants calculating the same metric include the same costs, hours and time base. Without common component definitions, a benchmark comparison measures accounting differences instead of maintenance performance.

A capital expenditure buys an asset or materially extends its life or capacity and is depreciated over years. Maintenance expense restores an asset to its existing capability and hits the current operating budget. Misclassifying rebuilds distorts maintenance cost metrics.

Availability is run time ÷ planned production time. Performance is actual output ÷ the output the ideal cycle rate would have produced in that run time. Quality is good units ÷ total units produced, so the product of the three is the share of planned time that made good product at rated speed.

TEEP = OEE × Utilization, where Utilization is planned production time ÷ all calendar time. OEE grades you against the hours you planned to run; TEEP grades you against every hour on the clock, exposing idle capacity.

Breakdowns and setup or changeover time reduce Availability. Idling, minor stops and reduced running speed reduce Performance. Process defects and startup rejects reduce Quality, and classifying a loss correctly tells you whether maintenance, operations or the process owns the fix.

Cpk = the smaller of (USL − mean) and (mean − LSL), divided by three standard deviations of the process. It measures how comfortably the process fits inside its specification while penalizing off-center running; a higher Cpk means fewer out-of-spec units without loosening the spec.

Usually nothing, because buffer stock or spare capacity absorbs the outage before it reaches the constraint. An hour lost on the constraint, by contrast, is an hour lost by the whole plant, which is why reliability effort and spares are aimed at the constraint first.

Technical review and approval of the change, then updates to drawings, operating procedures, PM tasks and spare parts, plus training for the people who will run and maintain it. Skipping the documentation step leaves maintenance tactics aimed at the old configuration.

MTBF = total operating time ÷ number of failures in that period. It applies to repairable items; the equivalent for non-repairable items is Mean Time To Failure. Where the failure rate is constant, λ = 1 ÷ MTBF and reliability over a period t is R(t) = e^(−λt).

MTTR = total time to repair or replace ÷ number of repair events; it is the maintainability measure. In its strict sense it counts active restoration time — diagnosis, repair and testing — whereas Mean Downtime (MDT) also includes administrative and logistics delays such as reporting, waiting for parts and lock-out, so MDT is always at least as long as MTTR.

Inherent availability Ai = MTBF ÷ (MTBF + MTTR) uses only active repair time, so it reflects the reliability and maintainability designed into the asset. Operational availability Ao = MTBM ÷ (MTBM + MDT) counts every maintenance event and all downtime including logistics delays, so it is what the plant actually experiences.

The curve has three regions: early-life failures with a decreasing rate caused by manufacturing or installation defects, a useful-life period with a roughly constant random rate, and a wear-out region where the rate rises with age. Replacing on a fixed interval only helps where age drives failure; in the other two regions it resets the clock into infant mortality for no gain.

β below 1 means a decreasing failure rate (infant mortality), β near 1 means random failures at a constant rate, and β above 1 means an increasing failure rate (wear-out). Only β above 1 supports age-based replacement. Random failures call for condition monitoring or redesign instead.

Whether the failure is evident or hidden, and whether its consequence is safety and environmental, operational, or non-operational. The consequence category sets how much a task must achieve, and the task survives only if it is technically feasible and worth doing.

RPN = Severity × Occurrence × Detection, each scored on the same rating scale. Detection is rated high when the failure is hard to detect before it causes harm, so poor detectability inflates the RPN just as a severe or frequent failure does.

The P-F interval is the time from the moment a failure first becomes detectable (P) to functional failure (F). Set the inspection interval to a fraction of it, commonly half or less, so at least one inspection lands inside the warning window and leaves time to plan the repair.

The consequence of failure — safety, environment, production loss, repair cost, quality — usually weighted by the likelihood of failure. The resulting ranking drives PM depth, spare-part stocking levels, and which assets earn a full RCM or FMEA study.

Shaft alignment and rotor balancing to defined tolerances, controlled fastener torque, correct lubricant type and fill level, and contamination control — executed to specification instead of 'good enough'. Together they remove the installation defects that cause infant-mortality failures after a rebuild.

Vibration analysis finds rotating faults such as unbalance, misalignment and bearing wear, while airborne ultrasound catches pressure leaks and the earliest stage of bearing distress. Infrared thermography finds loose or overloaded electrical connections and thermal hot spots. Oil analysis finds wear metals, contamination and lubricant degradation.

A task-by-task comparison of the competencies the work requires against the competencies the crew actually holds. The resulting gap list becomes the training plan, the hiring plan, and the justification for bringing in specialist support.

Responsible does the work, Accountable owns the outcome and signs it off, Consulted gives input before the decision, and Informed is told after it. Each task should carry exactly one Accountable name, or ownership quietly disappears.

Area crews are dedicated to one production unit, so they know the equipment and respond quickly but duplicate scarce specialist skills. A central crew pools those skills and levels workload but has less process familiarity. Many plants run a hybrid of the two.

Because the technical change is the easy half. People must understand the reason, see leaders using the new process themselves, and be measured on it. The plan defines the case for change, the communication, the training, and how new behavior gets reinforced.

A documented training and qualification path so one technician can safely finish a job that would otherwise need two crafts, and usually a pay structure that recognizes the added skills. The payoff is less waiting, travel and crew size on routine work.

Operators take on routine cleaning, inspection, lubrication and simple adjustments on their own equipment and report defects early — the TPM term is autonomous maintenance. It multiplies inspection coverage far beyond what the maintenance crew can reach and builds operator ownership of asset condition.

Convert undocumented know-how into standard job plans, procedures and failure histories, then run structured mentoring or apprenticeship so the successor performs the work before the expert leaves. Succession planning identifies the critical roles first.

Planning answers what, how and with what: scope, procedure, parts, tools, crafts and estimated hours. Scheduling answers who and when, committing named resources to a time window agreed with operations. Planning removes obstacles; scheduling allocates time.

Schedule compliance = scheduled work orders or hours completed as scheduled ÷ total work orders or hours on the frozen schedule × 100. Jobs added after the schedule was frozen were never on it, so they cannot count as compliant; they show up instead as reactive work, the SMRP metric for work that interrupts the weekly schedule.

PM compliance = PM work orders completed by their due date or within the organization's preset on-time criterion ÷ PM work orders due in the period × 100. Many plants allow a grace window tied to the PM interval, such as the common '10% rule'; a PM finished after that window counts as a miss, not a late pass.

Travel, waiting for parts, permits or instructions, hunting for information and attending meetings. Wrench time is the share of paid craft hours spent actively working at the job site, which is why better planning and kitting raise it more than working faster does.

In crew-weeks: total estimated craft hours of identified work ÷ the crew's available craft hours per week. Counting open work orders instead hides whether the queue holds twenty small jobs or twenty turnaround-sized ones.

Total backlog is every approved open job. Ready backlog is only the work that is fully planned with parts, tools and permits available, so it could go on next week's schedule. Schedulers build the weekly schedule from ready backlog.

Only when the job went through a formal planning process — scope, procedure, parts and tools defined — before the crew started. Percent planned work = craft hours on planned jobs ÷ total craft hours × 100; work that was merely scheduled, or planned after it began, does not count.

Emergency jobs pre-empt the frozen schedule, destroy wrench time while crews wait for parts and information, and force premium-priced expedited purchases and overtime. A rising share of emergency work therefore signals that the proactive program is losing ground, not just that equipment is aging.

Scope and task steps, safety requirements and permits, the crafts and estimated hours, parts and materials with storeroom locations, special tools or rented equipment, and the drawings or procedures needed. Repeatable jobs are saved as standard job plans and reused.

Priority should combine asset criticality with the consequence and urgency of the failure — safety and environmental risk first, then production impact, then cost. Letting whoever calls loudest set priority turns the schedule into a reaction queue and starves critical work.

The storeroom picks and assembles every part and material for a specific job in advance and stages it so the crew starts with all of it on hand. It converts storeroom trips into wrench time and surfaces missing parts before the job is scheduled.

Turns = annual value of material issued from stores ÷ average value of stores inventory. Low turns mean cash is parked in slow-moving stock, but cutting critical spares purely to raise turns trades a small carrying cost for a large downtime risk.

The percentage of parts requests filled immediately from stock — the mirror image of the stock-out rate (stock-outs ÷ requests) that SMRP tracks. It is the counterweight to inventory turns: the storeroom balances carrying cost against the downtime a stockout on a critical spare would cause.

Pareto analysis of failure causes, MTBF and other reliability calculations, and any maintenance strategy or PM review. The closed work order is the only record of what actually failed and why, so free-text notes without codes leave the history unsearchable.

It is a fixed-window project with a frozen scope, long-lead materials, contract labor and a critical path. Scope must be locked well before execution, because late scope additions are the classic cause of turnaround cost and duration overruns.

The share of maintenance labor hours (or work orders) spent repeating a job already performed, typically a repeat repair of the same failure on the same asset. It is a direct read on work quality and on whether the original root cause was actually removed.

The asset register and hierarchy, automatic PM and PdM triggers, parts linked to equipment, labor and cost capture, and searchable failure history. Its reliability value depends entirely on the quality of the data entered at work-order close-out.