CMRP common mistakes
CMRP · Understand the traps and how to avoid them.
Treating availability, reliability, and maintainability as interchangeable and computing the wrong metric for a scenario, such as dividing MTBF by MTBF+MTTR when asked for inherent availability including logistics delay.
The formulas look similar and many candidates memorize only one version, so they default to the simplest ratio regardless of what the question asks.
What to do instead
Write the three definitions side by side before answering: reliability is the probability of surviving a stated interval, maintainability is the probability of restoring within a stated time, and availability is uptime divided by total time. Then identify which downtimes the stem includes (repair only, or repair plus waiting for parts and labor) and select the matching formula.
Choosing the most advanced predictive technology for every condition-monitoring question instead of matching the technology to the failure mode, such as recommending vibration analysis for a slow-developing bearing fault but also for a blocked filter or a loose electrical connection.
Predictive maintenance is marketed as universally superior, and candidates equate 'modern' with 'correct' without mapping the monitored parameter to the physical degradation mechanism.
What to do instead
Build a quick mapping in your notes: vibration for rotating imbalance, misalignment, and bearing defects; thermography for electrical resistance and friction heating; oil analysis for wear particles and lubricant chemistry; ultrasound for leaks and early bearing or arcing signals. When a stem names the failure mode, select the technology that directly detects that mechanism.
Applying RCM decision logic from memory and assigning a scheduled overhaul to a hidden failure with no redundancy, or defaulting every failure to condition-based maintenance regardless of consequence.
Candidates remember the RCM flowchart shape but skip the consequence questions, so they choose a task before establishing whether the failure is evident or hidden and what it affects.
What to do instead
Work RCM questions in order: define function and failure mode, state the consequence (safety, environmental, operational, non-operational), then ask whether the failure is evident or hidden. Hidden failures with no redundancy usually call for a failure-finding task; evident failures with a measurable degradation interval call for on-condition or scheduled restoration. Eliminate any option that applies a task without a stated consequence.
Mixing up the roles and sequencing of work order types, such as treating a standing or blanket work order as a proper tool for routine PM execution, or closing a work order before capturing failure codes and repair history.
The exam tests work management as a system, and candidates who have only seen one plant's process assume their local shortcuts are standard practice.
What to do instead
Learn the standard lifecycle: notification, prioritization, planning, scheduling, execution, completion with failure coding, and closeout with history capture. Standing work orders are acceptable only for recurring minor tasks with defined scope, and any work order without a failure code or repair record is incomplete. On scenario questions, reject options that bypass planning, scheduling, or history capture.
Confusing the planning and scheduling functions, for example answering that the planner assigns daily work to individual technicians or that the scheduler procures parts for upcoming jobs.
In smaller organizations one person often does both, so the distinction is blurred in candidates' experience even though the exam treats them as separate roles.
What to do instead
Anchor on the definitions: planning determines what work is needed, the scope, the labor hours, the materials, and the permits; scheduling decides when the work will be done and by whom, within the weekly schedule. The planner prepares the job so it can be executed without delay; the scheduler sequences it against available capacity. Any option that has the planner assigning daily tasks or the scheduler ordering parts is wrong.
Treating criticality analysis and FMEA as the same exercise, or ranking assets by replacement cost rather than by consequence of failure and likelihood.
Both tools produce risk rankings, and candidates default to the most familiar one or to a financial proxy that feels objective.
What to do instead
Separate the purposes: criticality analysis ranks assets or systems to allocate attention and resources, using consequence and likelihood across safety, environment, production, and cost; FMEA analyzes failure modes within a component or process to identify causes, effects, and controls. Replacement cost is one input at most and never the sole ranking criterion. Match the tool named in the stem to its stated purpose.
Answering safety and environmental questions with the hierarchy of controls inverted, such as relying on PPE or training as the primary control for a recognized hazard when elimination or engineering controls are feasible.
Candidates recall the list of controls but not the ranking, and workplace habit often treats PPE as the default solution.
What to do instead
Memorize the order: elimination, substitution, engineering controls, administrative controls, then PPE last. On scenario questions, choose the option that removes or isolates the hazard at the source. If the stem describes a hazard that could be engineered out, any answer centered on retraining or additional PPE is a distractor.
Confusing inventory management metrics and policies, such as using economic order quantity when the stem calls for a reorder point, or treating safety stock as cycle stock.
The formulas share inputs like demand and lead time, and candidates pick the familiar formula without checking what quantity the question asks for.
What to do instead
Define each quantity in one line before computing: EOQ minimizes ordering plus holding cost; reorder point equals demand during lead time plus safety stock; safety stock buffers demand and lead-time variability; cycle stock covers normal replenishment between orders. Read the stem's final question, decide which quantity it names, then apply only that formula.
Treating training as an event rather than a system, for example answering that a training session alone closes a skills gap without verifying competence or measuring on-the-job performance.
Compliance-driven training cultures equate attendance with competence, and candidates carry that assumption into questions about development programs.
What to do instead
Apply a results chain to every training question: needs analysis, learning objectives tied to tasks, delivery, knowledge and skills assessment, and verification of on-the-job performance with follow-up. An option that stops at course completion is incomplete. When the stem asks how to confirm a training program worked, choose the answer that measures workplace behavior or results, not attendance or satisfaction scores.
Choosing a continuous improvement tool by name rather than by problem type, such as applying DMAIC to a process that has no baseline data, or using a kaizen event to solve a problem that requires capital investment and design change.
Candidates learn the tool families separately and default to the most familiar acronym instead of diagnosing the problem first.
What to do instead
Match the tool to the situation: DMAIC for existing processes with data and a defined problem; DMADV or design for new processes; kaizen events for focused, rapid waste elimination within existing scope; value stream mapping to expose flow and lead-time waste; 5S for workplace organization; root cause analysis methods for a specific recurring defect. On scenario questions, identify whether the problem is variation, waste, design, or a one-off failure before selecting the tool.