Free · No account required

Parameterize the real-world
disruption layer of ReliaSim

Traditional models simulate what machines are designed to do. ReliaSim's interrupt construct models what actually happens — stochastic failures and repairs that dynamically override nominal behavior, exposing system performance invisible to steady-state analysis. ReliaStats gives you the tools to parameterize that disruption layer with precision — directly in your browser, no install required.

Launch Designer → See how it fits

Runs entirely in your browser — data never leaves your machine

The parameterization challenge

Accurate models start with accurate distributions

Every interrupt captures a real-world disruption — a stochastic event that overrides nominal machine behavior. Each is defined by two distributions: how long a node runs before the event occurs (TTF) and how long recovery takes (TTR). Getting these right is the difference between a simulation that reveals true system dynamics and one that misleads.

"Traditional methods of parameterizing interrupts require dedicated time and attention from an expert to manually manipulate the tape and use their knowledge to distill line event data into cause groups — each with unique TTF and TTR distributions. The process is time-consuming, expensive, and requires an expert."

— ReliaSim documentation on interrupt parameterization

📉

TTF controls interruption frequency

Time to Failure defines how long a node operates before an interrupt occurs — random breakdowns, scheduled downtime, wear-out conditions, or volume-based triggers.

🔧

TTR determines how disruptive failures are

Time to Repair defines how long a node is unavailable once interrupted — maintenance, resets, refills, or operator response. Together, TTF and TTR define the reliability profile.

🎯

ReliaStats puts the power in your hands

Design distributions from scratch, explore fitted parameters, or compare versions side by side — no expert required, no data uploaded.

The heart of ReliaStats

LEDS Data to Fitting Engine

The engine that turns raw Line Event Data (LEDS) into parameterized interrupt distributions — automatically, accurately, and without requiring an expert modeler.

The problem

Traditional interrupt parameterization requires dedicated time and attention from an expert. Manual distribution fitting from event logs is slow, expensive, and scarce — a hard bottleneck for anyone who wants a validated production model.

What the engine does

Take a raw event tape in. Get parameterized TTF and TTR distributions out — mapped to cause groups, ready for ReliaSim. The expertise is built into the engine, not rented from a consultant.

Input

LEDS

Line Event Data

Engine

Fitting Engine

auto cause-group + distribution fit

Output

Parameterized

TTF / TTR → ReliaSim

"The power to start with a raw tape and produce parameterized interrupts quickly, accurately, and easily has been placed in any user's hands."

Where ReliaStats fits

The key to the ReliaSim process

ReliaStats owns the data integration step — converting raw event history into parameterized interrupt distributions, then closing the loop by validating simulation results against the same history.

🗺️

Step 1

Build

Define the production line in ReliaSim and parameterize each machine's interrupt behavior from raw historian event data.

ReliaStats

✅

Step 2

Validate

Confirm simulation results match real line performance — to within 1% statistical accuracy. Distributions built from time series that faithfully capture real equipment behavior.

🚀

Step 3

Decide

Experiment with what-if scenarios and commit capital where the validated model says it belongs — without touching the line.

Four components, one workflow

From raw event data to validated ReliaSim parameters

Design and understand distributions for free. Fit raw historian data with ReliaStats. View and compare the results — also free.

Inside ReliaStats

Five tools — one workspace

Design, fit, view, and compare interrupt distributions — then validate that ReliaSim's output matches the historian you fit from. Four tools free, forever. Only historian fitting requires a ReliaStats subscription.

Interrupt Designer

Build a TTF + TTR pair from scratch. Explore what distribution choices mean in practice before committing to a fit.

Working / Failed state timeline over a configurable shift window
Box-and-whisker: p5, Q1, Median, Q3, p95
Switch timeline scenario: Mean, Median, Q1, Q3, or p95
PDF, CDF, and Survival curves for TTF and TTR
Availability % from Mean TTF and Mean TTR

ReliaStats

Interrupt Fitting

Upload raw historian event logs. ReliaStats automatically identifies failure and repair patterns, fits the best distribution for each interrupt, and exports a ReliaSim-ready file.

Import raw failure & repair event logs
Automatic cause group identification
Goodness-of-fit tests (K-S, Anderson-Darling)
Distribution ranking and model selection
Export fitted interrupt file for ReliaSim

Interrupt Viewer

Load any ReliaSim interrupt file and browse every interrupt. Click a row to inspect its uptime and downtime distributions — fast audit of a full model's parameterization.

All 8 ReliaSim distribution types supported
Box-and-whisker percentile strip per distribution
PDF, CDF, and Survival function views

Interrupt Comparison

Two interrupt files side by side. Differences color-coded by severity; curves overlaid for direct inspection.

Color-coded rows by % change
Overlaid PDF / CDF / Survival curves
Every numeric difference surfaced automatically

Interrupt Validation

Validate against history — within 1%. Each point is an interrupt; on the diagonal = the sim matches the historian. ReliaSim's output cycles back here for comparison — refine and re-run until every interrupt lands on the line.

Source vs. sim availability, interrupt by interrupt
95% PI / 99% CI bands flag outliers instantly
Closes the fit → simulate → validate loop

Distribution support

All 8 ReliaSim distribution types

Every distribution ReliaSim supports, parameterized exactly as ReliaSim stores them — including real-space mean/std for LogNormal.

Weibull

λ (scale), κ (shape), γ (location)

Shape κ<1 = infant mortality, κ=1 = random, κ>1 = wear-out. Most versatile reliability distribution.

Wear-out · Random failures

LogNormal

mean (x̄), std dev (s)

Right-skewed — most values small, occasional large outliers. Ideal for variable repair durations.

Repair times · Compounded delays

Normal

μ (mean), σ (std dev)

Symmetric around mean. Use carefully for TTF — allows negative values unless constrained.

Consistent human-driven processes

Exponential

λ (rate) · mean = 1/λ

Memoryless — failure probability doesn't increase with age. Constant failure rate.

Random independent failures

Uniform

min, max

Equal probability across a range. Useful when failures can occur anywhere but no single value is preferred.

Limited data · Bounded range

Triangular

min, max, mode

Simple 3-point estimate. Easy to communicate to non-statisticians when only expert estimates are available.

Expert estimates · Quick modeling

Johnson SU

γ, δ, ξ, λ

Highly flexible 4-parameter system. Handles skewed and heavy-tailed behavior that simpler distributions can't capture.

Complex empirical data

Fixed / Schedule

x (value)

Deterministic — always returns the same value. Models predictable scheduled events with zero variance.

Shift changes · Planned maintenance

Reliability Engineering Fundamentals

The Bathtub Curve

Every component follows a failure rate pattern over its lifetime. Identifying which phase your equipment is in determines the right distribution — and the right intervention.

↘

Infant Mortality

Weibull (κ < 1)

Defects, installation errors, and manufacturing variation cause early failures that decrease over time. The component is most likely to fail shortly after being put into service.

In the Designer, a Weibull with κ < 1 captures this behavior — the hazard function is highest at t=0 and falls as the weak units fail out.

→ Burn-in testing · Strict incoming QA · Early replacement schedules

→

Random Failure

Exponential · Weibull (κ = 1)

Failures occur independently of age — external shocks, operator error, voltage spikes, random environmental events. The failure rate is constant: knowing how long a component has run tells you nothing about when it will fail next.

This is the "useful life" phase. The Exponential distribution (memoryless property) is the canonical model.

→ Redundancy design · Condition monitoring · Failure mode analysis

↗

Wear-out

Weibull (κ > 1) · LogNormal

Fatigue, corrosion, erosion, and material degradation cause the failure rate to increase with age. Failures become more predictable — and preventable — as the component approaches end of life.

Weibull with κ > 1 fits wear-out well. κ = 2–3 is common for mechanical fatigue; higher κ reflects tighter failure clustering.

→ Scheduled replacement · Life-limit programs · Predictive maintenance

ReliaStats Designer lets you select and parameterize the exact distribution that matches your equipment's phase — then see the impact on availability before touching a single setting in ReliaSim.

Open the Designer →

The ReliaStats product family

From distribution design to full automated fitting

ReliaStats Designer gives any engineer the tools to work with interrupt distributions. ReliaStats adds automated fitting from raw historian event data — replacing the expert-intensive manual process with one that anyone can run.

Free now

ChiAha

ReliaStats Designer

Design, explore, and compare TTF/TTR distributions. Runs entirely in your browser with no data upload.

✓ Interrupt Designer (TTF + TTR pair)
✓ Shift cycle state timeline
✓ Box-and-whisker percentile view
✓ Distribution Explorer (single CSV)
✓ Parameter comparison (two CSVs)
✓ All 8 ReliaSim distribution types
✓ Availability computation

Launch Designer →

ChiAha

ReliaStats

Convert a raw historian event tape into parameterized ReliaSim interrupts — quickly, accurately, and without needing an expert.

✓ Everything in Free
◎ Import raw failure & repair event logs
◎ Automatic cause group identification
◎ Automated distribution fitting
◎ Goodness-of-fit tests (K-S, A-D)
◎ Model ranking and selection
◎ Export fitted CSV for ReliaSim

🔒 Fitting runs in the cloud on anonymized numeric data only — interrupt names and descriptions are never transmitted.

◎ Coming soon

ChiAha

ReliaSim

Discrete-event reliability simulation engine. Models machine performance using interrupt-driven TTF and TTR distributions — within 1% statistical accuracy.

✓ Discrete-event simulation engine
✓ Interrupt-based reliability modeling
✓ Availability, throughput, efficiency output
✓ Buffer trade-off analysis
✓ CSV parameter import from ReliaStats

Learn more →

"Before ChiAha, it would take me up to a month to develop a digital twin for a production line. ChiAha streamlines this process, making it faster and more straightforward while still delivering high-quality results."

— Tom Lange, Technology Optimization & Management LLC · 36 years, Procter & Gamble · Co-author, "High Accuracy Discrete Rate and Reliability Modeling" (Winter Simulation Conference 2020)

The underlying methodology has been validated by 300+ organizations across food & beverage, pharma, aerospace, defense, and energy since 1995. Current customers include Essity and Process Partners.

Parameterize the real-world
disruption layer of ReliaSim

Accurate models start with accurate distributions

TTF controls interruption frequency

TTR determines how disruptive failures are

ReliaStats puts the power in your hands

LEDS Data to Fitting Engine

The problem

What the engine does

The key to the ReliaSim process

From raw event data to validated ReliaSim parameters

Five tools — one workspace

Interrupt Designer

Interrupt Fitting

Interrupt Viewer

Interrupt Comparison

Interrupt Validation

All 8 ReliaSim distribution types

The Bathtub Curve

From distribution design to full automated fitting

Built for high-speed, high-volume operations

Start getting your Interrupt parameters right

Join our list

Parameterize the real-world disruption layer of ReliaSim

Accurate models start with accurate distributions

TTF controls interruption frequency

TTR determines how disruptive failures are

ReliaStats puts the power in your hands

LEDS Data to Fitting Engine

The problem

What the engine does

The key to the ReliaSim process

From raw event data to validated ReliaSim parameters

Five tools — one workspace

Interrupt Designer

Interrupt Fitting

Interrupt Viewer

Interrupt Comparison

Interrupt Validation

All 8 ReliaSim distribution types

The Bathtub Curve

From distribution design to full automated fitting

Built for high-speed, high-volume operations

Start getting your Interrupt parameters right

Join our list

Parameterize the real-world
disruption layer of ReliaSim