Patok Lab
Get access
ES

Discrete-event simulation Available

Ask what the line would do before you change it.

Entities through resources, queues and buffers: cycle times, setups, failures, shifts, batching, dispatch and release policies. Run it, replicate it with confidence intervals, compare two scenarios side by side, and let the optimizer search the policy — all from a model you can read and edit.

Get early access See the other benches
Patok Lab — a discrete-event model with its resources, queues and the results of a run

Model → run → replicate → compare → optimize

  1. 01

    Describe the flow

    Resources with cycle times, setups and failures; queues, buffers and blocking; shifts and batching; the release rule and the dispatch rule. Distributions where you have them: constant, triangular, normal, lognormal, exponential, Erlang, empirical.

    Patok Lab — the model form with resources, queues and policies
  2. 02

    One run is one seed

    A single run is a sample, not an answer. Replicate it and you get a mean and a confidence interval instead of a number that changes every time you press Run.

  3. 03

    Compare, then optimize

    Two scenarios in one call, with the differences that matter highlighted. When the question is a policy rather than a scenario, the optimizer searches it — seeded, so the answer reproduces.

  4. 04

    It is a capability like the others

    The model has a Code tab: the exact calculation that ran when you pressed Run. Results are artifacts with provenance, and they can go into a runs table, a report or a live app artifact that asks for new runs.

What is in the app today

  • A repeatable, seeded engine: the same inputs give the same answer
  • Replications with a mean and a confidence interval, not one lucky run
  • Scenario comparison in one call, and PSO for policy search
  • Release rules: immediate, EDD backward, offsets, CONWIP
  • Dispatch rules: FIFO, EDD, SPT, LPT, priority, critical ratio
  • A published factory-flow study reproduced bit-exact as a verification fixture
  • Every run a `simulation-result` artifact, with its assumptions listed

Before the spindle turns

  • This is discrete-event and queueing, not physics: no CFD, no FEA, no thermal solver.
  • A replica of a published study is a replica of its model, and every gap between the paper and the code is listed as an assumption on the run.
  • A simulation is as good as the times and the rules you put in it. The app will not invent them for you.

Typical starting points

  • One more machine, or one fewer?

    Model the cell as it runs today, then ask what the queue does when you add capacity — with a confidence interval, not a single run.

  • A release rule you argue about

    CONWIP against immediate release on the same instance, same seeds, side by side.

  • A clinic, a counter, a warehouse

    It is not only factories: anything where entities wait for resources behaves the same way.

  • A study you want to check

    Reproduce a published model, see where your numbers differ, and read the assumptions the replica had to make.

One workbench, six benches

Early access

Early access for people who run real processes — labs, engineering teams and shops.

Bring a customer drawing, a table of runs, a folder of micrographs or a part you spray, weld or mill. We want Patok Lab used on live work and told what breaks. The Community edition is free, with no time limit; macOS, Windows and Linux.

  • macOS, Windows and Linux — tell us which one you need first
  • Direct line to the founder during early access
  • Optional workshop for your group or team: bring a job, leave with an artifact
  1. 01Tell us what you would bringA drawing, a table of runs, a folder of micrographs, a part, a line.
  2. 02The build arrives by e-mailInstaller, a licence you paste once, and a short guide for the bench you came for.
  3. 03Reply with your fileWe look at it before we send it, and a person answers — not a form.

Get access Bring a drawing, some photos, a table or a part. Or just tell it.

Bringing it to a team, a shop or a university group? Ask for your team →

FAQ

Questions engineers and researchers ask first.

Do I need to know simulation to use it?

You need to know your process. The model asks for the things you already say out loud: how long a step takes, how often it fails, what the queue does when it is full, how work is released. The statistics of replication are handled for you.

Can I see the code?

Yes — every capability has a Code tab showing the exact calculation that ran when you pressed Run, and for capabilities you promote yourself you can edit it, test it in a separate interpreter and save it as a version.

Is it connected to the rest of Patok Lab?

Yes. Results are artifacts with provenance, they can become columns in a runs table, and a campaign can vary the model’s parameters the same way it varies a spray recipe.