ReliaSim + ReliaStats
ReliaSim + ReliaStats

Six steps from
production graph
to confident decision.

Graph
1
2
Parameterize
2A
Simulate
3
Validate
4
Experiment
5
Decide
6

What would you change in your production system?
How do you know the impact?

How do you increase production?

A bottling line — Filler, Capper, Labeler, Case Packer, Palletizer
Increase buffer?
Reduce Labeler Roll Change?
Add an automated Case Packer?

How do you decide?

Gut feel?
Spreadsheets?
Past experience?
Consultants?
Step 1

Click-connect what is.
Prepare for what if.

Draw your production line node by node.
Set units, rates, and conversions. Add buffers and mark decoupling points.

"It runs on my desktop. No integration, no IT project, no waiting." — Plant Manager, Food & Beverage

Duration: 90.00 Days Efficiency: 100.0% Perfect Production: 129,600 pallets
Buffer Source
Constraint Filler
Constraint Capper
Constraint Labeller
Constraint Case Packer
Constraint Palletizer
Buffer Sink
Steps 2 & 2A

Use your data to drive
the model's behavior.

Feed your historian data to ReliaStats to find the distributions that best fit each machine's failure behavior.

Historical Line Event Data — blocking and starving emerge from the interaction
Full line dynamics — five machines with blocking and starving

For systems without data, the Interrupt Designer guides your distribution selection based on your reliability process.

Interrupt Designer — select uptime and downtime distribution shapes
Interrupt Designer — TTF and TTR distribution shape selection
Step 3

Simulate your baseline.

Evaluate your system with eyes wide open to its range of outcomes.

Any duration Random seed analysis Single run Multiple runs
Constraint timeline
Constraint timeline
Throughput distribution
Throughput histogram
Production vs. ideal
Production through time
15 min
To build & run
1,200×
Faster than traditional
Step 4

Validate against history.
Within 1%.

Model vs. historian — simulated availability vs. actual
Bivariate validation — simulated vs actual availability within 1%

Each point is an interrupt. On the diagonal = model matches reality.

"The only way to predict this accurately is through simulation."
— Tom Lange, 36 years Procter & Gamble
Step 5

Experiment & Compare.

Labeler Misalignment
Infrequent, longer stops
Loss6.78%
Higher direct loss
Filler Micro Stop
Frequent, short stops
Loss6.67%
Lower direct loss

Similar direct losses. Your Pareto says: similar priority.
Eliminate each one and measure the whole system.

Step 5

Gain ≠ Loss.

Efficiency Gain/Loss — evaluate each interrupt, one at a time
Gain/Loss chart — blue bars are loss, orange bars are gain
Labeler Misalignment
0.75×
Loss6.78%
Gain5.10%
Filler Micro Stop
1.2×
Loss6.67%
Gain7.97%
Step 5

Frequent short stops
cascade.

120 one-minute jams stress the entire system.
Losses appear under other machines' names — invisible in your Loss Tree.

Full line dynamics — blocking and starving are emergent
Five machines — blocking and starving emerge from interaction
"Those 120 one-minute interruptions create cascading problems that don't show up under the original problem's name. You often recover 180–220 minutes — significantly more than the 120 minutes of downtime."
— Tom Lange, 36 years Procter & Gamble
Step 6

Make confident decisions.

No expensive consultants required. The expertise is built into the system.
Here are the decisions your validated model answers.

🔧

Which failure first?

Rank by system gain, not downtime.

⚡

Faster or slower?

Faster = more product, more failures.

📦

How much buffer?

Protect the bottleneck — but how much?

🔁

Second machine?

Redundancy vs. reliability improvement.

🔄

Manual or auto resupply?

True cost is often 3× what downtime says.

🎯

Your question here.

Every line has its own version.

ReliaSim + ReliaStats

See the impact of change
before you make it.

~1%
OEE accuracy
15 min
Model construction
1,200×
Faster than traditional
300+
Organizations

"It runs on my desktop. No integration, no IT project, no waiting."

— Plant Manager, Food & Beverage

Tom Lange & Andrew Siprelle — Executive Platforms Blueprint Podcast

ChiAha — Advanced Analytics & Automation Group · chiaha.com