Short, precise definitions of the terms used in production line simulation, OEE and reliability work. Each entry links to the guide that covers the term in depth, and to the matching ReliaSim documentation page where one exists.
A · B · C · D · F · H · I · L · M · O · R · S · T · V · W
A
Accumulating conveyor
A conveyor that moves material between process steps and can also hold it when downstream flow stops, so units queue on the conveyor itself. It combines transport delay and in-line buffering in a single element, with capacity set by its length and how many units fit per unit of length.
Learn more: Accumulating Conveyors · Choosing Between Conveyor Types
Agent-based simulation
A dynamic simulation method in which system behavior emerges from many individual agents, such as buyers, people or vehicles, each following its own decision rules and interacting with the others. It fits systems driven by individual decisions and interaction, where discrete event simulation fits systems better described as a process of defined steps and queues.
Learn more: Simulation methodologies compared
Asset efficiency
Output as a share of what a line could produce at rated speed over every minute of the modeled period, scheduled or not. It is the measure ReliaSim reports as Efficiency. Because its denominator is calendar time, it behaves like TEEP rather than OEE: on a plant running around the clock the two land close together, and on a plant running fewer shifts the gap equals the unscheduled time.
Learn more: Asset Efficiency vs OEE vs TEEP · Efficiency and OEE
Availability
The share of time a machine or system is able to run. In steady state, availability = MTBF ÷ (MTBF + MTTR), so it depends on both how often failures occur and how long repairs take. OEE Availability is narrower: run time ÷ planned production time, so setups and changeovers count against it as well as failures.
Learn more: Availability vs Reliability · Efficiency and OEE
B
Blocked
The state of a machine that is ready and able to run but has finished a unit and has nowhere to put it, because the buffer or conveyor downstream is full. The cause of blocking is downstream, and blocked time is not the machine's own downtime.
Learn more: Starved vs Blocked
Bottleneck
The resource that limits a production system's throughput. On a line with downtime it is often not the slowest rated machine: it is set by effective rate under real failure behavior, and it shows up as the station that is least often starved or blocked. It can also move with downtime, product mix and schedule. Also called the constraint.
Learn more: Find the Real Bottleneck on a Production Line
Buffer
Storage between process steps, such as an accumulation table, tank, silo or accumulating conveyor, that lets the machines on either side keep running while the other is stopped or slower. A buffer protects the downstream machine only while it holds material, and the upstream machine only while it has room, so a stop that outlasts the buffer passes straight through.
Learn more: How Big Should a Buffer Be? · Buffers
Buffer sizing
Choosing a buffer's capacity. Line efficiency typically rises steeply with capacity and then flattens, so the useful size covers the stop durations that actually occur, judged on their distribution rather than the average, and is weighed against the cost of the storage and against fixing the downtime it absorbs.
Learn more: How Big Should a Buffer Be? · Buffer Trade-off Results
Buffer tradeoff analysis
An experiment on a validated line model that sweeps a buffer's capacity and reads line efficiency against it, with interrupts selectively removed. It finds the knee where extra storage stops buying throughput, and compares adding storage with fixing the downtime that storage would absorb.
Learn more: OEE Improvement Tradeoffs · Buffer Trade-off Results
C
Capacity
The rate a line could produce. Rated capacity is what the machines deliver while running, fed and unblocked, which on a series line is set by the slowest machine's rated speed; effective capacity subtracts planned losses such as changeovers. Capacity is not throughput, which is what the line actually delivers.
Learn more: Throughput vs Capacity
Capacity planning simulation
Modeling a production line or plant with its rates, storage, changeovers and per-failure-mode downtime, then running it forward to see how much it produces under a given demand and product mix. It answers whether the line will make the volume before capital or overtime is committed.
Learn more: Capacity Planning Simulation
Constraint
In the theory of constraints, the resource that limits the throughput of the whole system, also called the bottleneck. In a ReliaSim model, a constraint is also a node type: a rate-limited step, such as a machine, where a maximum rate and interrupts are defined.
Learn more: Theory of Constraints Simulation · Constraints
Continuous simulation
A dynamic simulation method that treats state variables such as a tank level, a temperature or a pressure as changing continuously over time. The model is usually a set of differential equations, advanced by a numerical integrator in small time steps.
Learn more: Simulation methodologies compared
Converter
A ReliaSim node that combines a rate-limited operation with a change of units in one step, for example bottles into cases, and can accept more than one input stream. Like a constraint, it can carry interrupts, so it starves and blocks its neighbors when it stops.
Learn more: Converters · Conversions
D
Decoupling buffer
Inventory placed deliberately at a chosen point so that variability on one side of it does not propagate to the other. Lean and theory of constraints practice both converge on it, and demand-driven planning formalizes the same idea as a strategic decoupling point.
Learn more: How Big Should a Buffer Be?
Digital twin
A virtual representation of one specific physical asset or process that stays connected to it, so data flows automatically from the real system into the model and, in the strict definition, back again. A model refreshed by hand from the line's data is a simulation model (a digital model), and one fed automatically in one direction only is a digital shadow.
Learn more: Digital Twin vs Simulation
Discrete event simulation (DES)
A dynamic simulation method that assumes the system's state changes only at distinct instants, such as an arrival, a start or a finish, and jumps the clock from one event to the next. Each unit is typically modeled as an entity, which suits job shops, queues and systems where unit identity matters, but the number of events grows with the number of units on high-speed lines.
Learn more: Simulation methodologies compared · Discrete Rate vs Discrete Event Simulation
Discrete rate simulation (DRS)
A simulation method that models material as flow at rates that stay constant between events, so buffer levels change linearly and the moment a buffer will fill or empty can be calculated and scheduled; an event is a change in rate, not the passage of a unit. It sits between discrete event and continuous simulation. Andrew Siprelle created it in 1990 as bulk flow simulation, and it was renamed discrete rate simulation in the late 2000s.
Learn more: Discrete Rate vs Discrete Event Simulation · Simulation methodologies compared · ReliaSim Overview
Downtime Pareto
A bar chart of downtime causes sorted from largest to smallest, usually with a cumulative line, so the few causes behind most of the loss stand out. It can rank causes by number of stops, by downtime minutes or by lost production, and the three rankings rarely agree.
Learn more: Downtime Pareto: Frequency vs Duration vs Lost Throughput
Drum-buffer-rope (DBR)
The theory of constraints method for pacing a line. The drum is the constraint's pace, which sets the rate for the line; the buffer is stock, measured in time, that protects the constraint from upstream stoppages; the rope ties material release to the constraint so the rest of the line does not overproduce.
Learn more: Theory of Constraints Simulation
F
Failure mode
A distinct way a machine stops, such as a jam, a sensor fault or a component failure, recorded as a cause code in stop-event data. Each failure mode has its own time-to-failure and time-to-repair behavior, so each is fitted separately; pooling them hides the shape of every one. In ReliaSim, each failure mode is modeled as an interrupt.
Learn more: Downtime Data Analysis · Interrupts and Constraints
H
Hidden factory
The production a plant could get from equipment it already owns without buying anything, a term that traces back to Armand Feigenbaum. OEE cannot size it, because most of it sits in unscheduled time, which OEE excludes by design; a calendar-time measure such as TEEP or Asset Efficiency can.
Learn more: The Hidden Factory: How Big Is Yours?
I
Interrupt
ReliaSim's term for an event that makes a node unavailable, such as a random failure, a scheduled break, a wear-based stop or a volume-based changeover. Each interrupt has a time-to-failure distribution, which sets when it occurs, and a time-to-repair distribution, which sets how long it lasts; its type decides whether time to failure counts wall-clock time, uptime or processed volume.
Learn more: Interrupts and Constraints · Interrupt Types
L
Line event data (LEDS)
The stop-by-stop event record of a production line: for each stop, the machine, the cause and when it started and ended. It is the raw record interrupt distributions are fitted from, as opposed to rolled-up shift or daily totals.
Learn more: Downtime Data Analysis · Bottling Line Demos
Line OEE
OEE measured at the end of the line: good units out, at the line's rated speed, over the line's planned production time. It is not the average of machine OEEs, because on a series line losses pass from machine to machine; five machines at 85% with no buffering give a line near 44%. Buffers raise line OEE, but never above the weakest machine.
Learn more: Line OEE vs Machine OEE
Loss/gain analysis
An experiment on a validated line model that removes each interrupt in turn, re-runs the model and measures the production recovered. It ranks losses by recoverable throughput instead of minutes lost, and shows where buffers or cascades make the two rankings disagree, including interrupts whose removal costs output because they prevent a more disruptive one.
Learn more: OEE Improvement Tradeoffs · Gain/Loss Results
Loss tree
A spreadsheet breakdown of lost production by category and cause that adds losses as if they were independent. It is transparent and usually right about the total, but it cannot see interaction, such as where a stop sits relative to buffers and the constraint, so it misranks improvement projects and overstates what fixing a loss will return.
Learn more: Loss Tree, AI, or Simulation?
M
Machine OEE
OEE calculated for a single machine from its own run time, speed and good count. Time the machine spends starved or blocked by its neighbors is often excluded from its own figure, which is why every machine's OEE can look healthy while line OEE is low.
Learn more: Line OEE vs Machine OEE
Micro-stop
A very short stoppage, such as a jam cleared by the operator or a fault reset at the panel, often too short to be logged as downtime. In OEE, minor stops count as Performance loss, so they look like a machine running slightly slow. Because they keep buffers from refilling, frequent micro-stops can cost more throughput than rare breakdowns.
Learn more: Micro Stops vs Breakdowns
Monte Carlo simulation
A static method that repeatedly samples uncertain inputs from probability distributions, computes a result for each sample and reports the distribution of results. It has no simulated clock and carries no state forward, so it cannot represent buffers filling, machines starving or how long stops last.
Learn more: Monte Carlo vs Discrete Rate Simulation · Simulation methodologies compared
MTBF (mean time between failures)
The average running time between stops. It summarizes what already happened, so it is an output of a line model, not an input: a model takes time-to-failure distributions per failure mode and reports MTBF as a result. A machine's MTBF also changes when the line changes, because a starved or blocked machine is not running and so is not failing.
Learn more: MTBF vs MTTR: Outputs, Not Inputs
MTTR (mean time to repair)
The average duration of a stop, from failure until the machine is running again. Like MTBF, it is a reporting measure and a model output rather than an input; two machines with the same MTTR can affect a line very differently, so models take the full time-to-repair distribution of each failure mode.
Learn more: MTBF vs MTTR: Outputs, Not Inputs
O
OEE (overall equipment effectiveness)
Availability × Performance × Quality, where Availability = run time ÷ planned production time, Performance = ideal cycle time × total count ÷ run time, and Quality = good count ÷ total count. The product equals good count × ideal cycle time ÷ planned production time. Because it counts only planned time, OEE grades how well the plan was executed.
Learn more: OEE Calculation: Line OEE vs Machine OEE · Efficiency and OEE
OEE simulation
Building a model of a production line from its rates, buffers and per-failure-mode interrupt data, validating it against measured OEE, and running it forward to predict how line OEE responds to a change before the change is made.
Learn more: OEE Simulation
R
RAM analysis
The study of a system's reliability, availability and maintainability: how often its equipment fails, how quickly it is restored, and what share of the time, or of planned production, the system delivers as a result. A RAM study is typically used at the design or capital-planning stage to test whether a design will meet its production target.
Learn more: RAM Analysis
Reliability
The probability that a machine or system runs for a stated period without failing. It is distinct from availability: a machine that stops about once an hour for five minutes is roughly 92% available, yet almost never completes an eight-hour shift without a stop.
Learn more: Availability vs Reliability
Reliability block diagram (RBD)
A diagram of a system as blocks, each with its own reliability or availability, arranged in series (every block needed) or in parallel (at least one needed), from which system reliability or availability is calculated assuming independent failures. It produces availability, not throughput: it cannot represent rates or storage, so on a line with buffers a series RBD overstates the loss.
Learn more: Reliability Block Diagram vs Line Simulation
S
Six big losses
The TPM classification of equipment losses: equipment failure and setup and adjustment (Availability), idling and minor stops and reduced speed (Performance), and process defects and reduced yield at startup (Quality). They give a complete inventory of where production goes, but ranking them by minutes lost ranks how often losses occur, not what they cost.
Learn more: Which of the Six Big Losses to Fix First
Starved
The state of a machine that is ready and able to run but has nothing to work on, because the buffer or conveyor upstream is empty. The cause of starvation is upstream, and starved time is not the machine's own downtime.
Learn more: Starved vs Blocked
T
TEEP (total effective equipment performance)
OEE × Utilization, where Utilization is the share of calendar time scheduled for production. Its denominator is every hour the asset exists, staffed or not, so a line can post excellent OEE and mediocre TEEP at the same time; TEEP answers the capacity question that OEE excludes by design.
Learn more: Asset Efficiency vs OEE vs TEEP
Theory of constraints (TOC)
Eliyahu Goldratt's management approach, built on the premise that a system's output is limited by its constraint. It improves throughput through five focusing steps: identify the constraint, exploit it, subordinate everything else to it, elevate it, then repeat when the constraint moves.
Learn more: Theory of Constraints Simulation
Throughput
The rate of good units a line actually delivers, after unplanned stops, slow running, rejects and time spent starved or blocked. For a single machine, throughput rate = rated rate × Availability × Performance × Quality; for a line the same formula holds with line OEE, which has to be measured or simulated.
Learn more: Throughput vs Capacity
Time to failure (TTF)
The running time of one individual run before a stop. Its distribution, fitted per failure mode, describes how runs vary, and MTBF is only its mean. A simulation samples TTF for each interrupt, counted against wall-clock time, uptime or processed volume depending on the interrupt type.
Learn more: MTBF vs MTTR: Outputs, Not Inputs · Time to Failure and Time to Repair
Time to repair (TTR)
The duration of one individual stop, from failure until the machine is running again. Repair times are usually right-skewed, with most stops quick and a few long, so a fitted distribution such as the lognormal says far more about a failure mode than its mean, MTTR.
Learn more: MTBF vs MTTR: Outputs, Not Inputs · Time to Failure and Time to Repair
V
Value stream map (VSM)
A Lean diagram of the material and information flow for a product family, showing the process steps, the inventory between them and summary data such as cycle time, uptime and lead time. Its numbers are averages and snapshots with no clock, so it cannot show buffers filling and draining or the constraint moving between machines.
Learn more: Value Stream Map Simulation
W
Weibull distribution
A two-parameter distribution, with shape κ and scale λ, widely used for time-to-failure data. It takes only positive values, and its shape parameter lets it represent decreasing, constant or increasing failure rates. The scale λ is the running time by which 63.2% of runs have ended, and the mean is λ · Γ(1 + 1/κ).
Learn more: Weibull Analysis of Downtime Data · Distributions
Weibull shape parameter
The Weibull parameter κ (often written β) that sets how the failure rate changes with running time: below 1 it falls, pointing to early-life or restart-related failures; at 1 it is constant, which is the exponential distribution; above 1 it rises, indicating wear-out. A single fit across pooled failure modes can look random when no individual mode is, so each mode is fitted separately.
Learn more: Weibull Analysis of Downtime Data · Distributions
World-class OEE
The widely quoted 85% OEE benchmark from Seiichi Nakajima: 90% Availability × 95% Performance × 99% Quality. It is a single-machine benchmark; five world-class machines in series with no buffering give a line OEE near 44%, so a line's realistic ceiling depends on its topology and buffering.
Learn more: World-Class OEE: Is 85% Achievable on Your Line?