CRPE Engineer Registration key terms

The probability that an item performs a required function under stated conditions for a stated period of time. It is expressed as a probability between zero and one and is always tied to defined operating conditions and a mission duration.

The probability that a failed item can be restored to a specified condition within a stated time when maintenance is performed under stated conditions using prescribed procedures and resources. It is a design characteristic measured by maintenance time distributions.

The probability that an item is in a functioning state at a random point in time, or the fraction of time it is operational. It combines how often failures occur with how quickly they are repaired.

The conditional probability that an item fails in the next unit of time given that it has survived to the start of that interval, usually denoted lambda. Often reported as failures per million hours or per billion hours.

The average operating time between consecutive failures of a repairable item. For a constant failure rate it equals the reciprocal of the failure rate and is used as a summary measure of field reliability.

The average time required to restore a failed item to operation, including fault detection, isolation, disassembly, replacement, reassembly, and verification. It is the central parameter of maintainability models and availability calculations.

The steady-state availability computed from mean time between failures and mean time to repair alone, assuming only corrective maintenance and no logistics delay. It represents the availability achievable from the design and is a common contractual metric.

The availability actually achieved in the field, including corrective and preventive maintenance time plus administrative and logistics downtime. It is always lower than inherent availability because it accounts for supply and support delays.

A graphical model showing three life phases: decreasing infant mortality failures, a roughly constant useful-life failure rate, and increasing wear-out failures. It guides burn-in decisions, spare parts planning, and replacement intervals.

A flexible life distribution with shape and scale parameters that models increasing, constant, or decreasing hazard rates. The shape parameter indicates whether failures are infant mortality, random, or wear-out in nature.

The life distribution with a constant failure rate, meaning the probability of failure in the next interval does not depend on age. It is the basis for simple reliability predictions and Poisson failure-count models.

A statistical method that chooses parameter values making the observed data most probable. It is widely used to fit life distributions and is preferred for censored data because it uses partial information from unfailed units.

Observations where the exact failure time is unknown, either because testing stopped first (right censoring) or because the item already had failed before observation began. Ignoring censoring biases life estimates.

A graph of life data transformed so that a chosen distribution plots as a straight line. Departures from linearity reveal the wrong distribution choice or mixed failure modes, and the slope estimates the shape parameter.

A range of values, computed from sample data, that contains the true parameter with a stated probability over repeated sampling. Reliability intervals are usually one-sided because only the lower bound is of practical interest.

Testing at stress levels above normal use, such as higher temperature or voltage, to provoke failures sooner. A physics-based model then extrapolates results back to use conditions, reducing test duration.

The ratio of life or failure rate at use stress to that at accelerated stress. It converts accelerated test time into equivalent use time and must be validated for the failure mechanism being exercised.

An empirical relationship in which reaction rate depends exponentially on inverse absolute temperature. It is the standard acceleration model for temperature-driven failure mechanisms such as chemical degradation and electromigration.

A test designed to show, with stated statistical confidence, that reliability meets a specified requirement. It uses a defined accept-reject rule and a test time derived from the requirement and risk levels.

The probability that a system meeting its reliability requirement is nevertheless rejected by the demonstration test. It is the alpha error and is balanced against consumer's risk when designing the test plan.

The probability that a system failing to meet its reliability requirement is nevertheless accepted by the demonstration test. It is the beta error and drives the test duration needed to protect the customer.

The systematic improvement of reliability over a development program as design defects are found and corrected. It is tracked on growth plots and modeled to forecast whether the requirement will be reached on schedule.

An empirical growth model in which cumulative failure rate plotted against cumulative test time on log-log axes forms a straight line. Its slope, the growth rate, indicates how effectively corrective actions reduce failures.

A reliability growth process in which failures found in testing are analyzed to root cause, corrected in design, and verified. Fix effectiveness and delayed fixes determine how much reliability actually improves.

A screening period in which items operate under stress before delivery so that early-life defects fail in the factory rather than in the field. Its duration is set to remove infant mortality without consuming useful life.

Applying thermal cycling, vibration, or other stresses to production items to precipitate latent manufacturing defects. It differs from burn-in in that it targets workmanship flaws rather than design weaknesses.

A qualitative test that steps stress up until failures occur and then reduces stress to find operating and destruct limits. It exposes design weaknesses quickly but does not directly quantify field life.

A preliminary reliability estimate summing component failure rates from generic reference data, usually assuming constant failure rates and series dependency. It is fast and useful early in design when detail is limited.

A detailed reliability prediction that adjusts each component's generic failure rate for its actual applied stress, temperature, and quality level. It is more accurate than parts count but requires a complete circuit and thermal design.

Operating a component below its rated electrical, thermal, or mechanical stress to reduce its failure rate and extend life. Derating guidelines specify maximum stress ratios for each part category.

Providing more than one means of performing a function so that failure of one element does not cause loss of the function. It may be active, standby, or voting and raises reliability at the cost of complexity and weight.

A single shared event or condition that disables multiple supposedly independent redundant elements at once, such as a shared power supply or a common manufacturing defect. It defeats redundancy unless explicitly defended against.

The fraction of total failure rate attributable to common cause events affecting all channels of a redundant group. A nonzero beta factor caps the reliability benefit that redundancy can deliver.

A bottom-up technique that lists each item's failure modes, traces their local and system effects, and ranks severity and criticality. It identifies single points of failure and drives design and test priorities.

A ranking score formed by multiplying severity, occurrence, and detection ratings in failure mode analysis. Higher scores flag modes for corrective action, though the product has known limitations in interpretation.

A top-down deductive method that starts from an undesired top event and traces combinations of lower-level events through AND and OR gates. It yields minimal cut sets and quantitative probability estimates.

A smallest combination of basic events whose simultaneous occurrence causes the top event. Single-element cut sets reveal single points of failure and are the primary targets for design improvement.

A success-oriented model that represents the system as blocks whose series or parallel arrangement mirrors functional dependency. It computes system reliability from element reliabilities and complements fault tree analysis.

A reliability model in which every element must succeed for the system to succeed, so system reliability is the product of element reliabilities. Adding elements always lowers system reliability in this arrangement.

A reliability model in which the system succeeds if at least one element succeeds, so system reliability is one minus the product of element unreliabilities. It represents active redundancy in a block diagram.

A component or interface whose failure alone causes loss of the entire function or a safety consequence. Such items are identified in analyses and eliminated or mitigated by redundancy or design change.

The likelihood that a required human action is performed incorrectly, estimated from task analysis, simulator data, or performance-shaping factors. It feeds human reliability analysis and risk models alongside hardware failure rates.

A condition such as time pressure, training, fatigue, or interface quality that influences the likelihood of human error. These factors adjust nominal error probabilities in human reliability analysis.

A structured method for identifying human tasks that affect safety or availability and quantifying the chance of error. It includes task analysis, error identification, and modeling of recovery opportunities.

A closed-loop management process for recording failures, analyzing root causes, implementing corrective actions, and verifying their effectiveness. It supplies the data used for growth tracking and field reliability assessment.

A structured investigation to find the underlying physical or process reason a failure occurred, rather than the immediate symptom. Effective corrective action depends on identifying the true root cause.

A document defining the reliability tasks, responsibilities, schedule, and deliverables for a program. It tailors standards to contract requirements and links each task to design milestones and reviews.

A description of the environmental and operational conditions an item experiences over its life, including durations, cycles, and stress levels. Reliability requirements and predictions are meaningless without it.

The amount by which a design's capability exceeds the expected worst-case stress. Adequate margin protects against variability, wear, and unknown loads, and is verified through analysis and margin testing.