Home/Services

Simulation and operations research services

We build models of production lines, plants and supply chains from your own data. Each model has to reproduce what the plant actually did before we use it to answer anything. Pick the question closest to yours below.

Engagements

Start from the decision in front of you

Each page below covers one kind of engagement in detail: what we model, the data it takes and what you get back.

What's in the work

Methods we bring to each engagement

The mix depends on the question. Most projects use several of these together, and the tools behind them are the same ones we sell.

OEE and throughput simulation

Capacity modeling, bottleneck identification and what-if analysis on the line as it really runs, with stops, starving and blocking included.

Theory of constraints and lean

Finding the constraint that actually governs output. On a line with frequent stops it often moves as failures come and go.

Verification and validation

Comparing the model against your historian data, failure mode by failure mode, before anyone trusts what it predicts.

Discrete rate modeling

Line flow modeled as rates that change when machines stop and start, so a year on a high-speed line simulates in seconds.

Feasibility and sensitivity analysis

Testing a configuration before you buy it, and finding out which uncertain inputs actually change the answer.

Practical AI

AI assistants that call the simulation engine and quote its results, so plant teams can ask what-if questions in plain language.

How we work

Three ways to work with us

Engagements run on the same engines as our products, so the model doesn't disappear when the project ends. The manufacturing simulation page walks through how an engagement runs, step by step.

1

We build the model

We scope it to your decision, build it from your data, validate it against history and hand over the answer with the evidence behind it.

2

We train your analysts

Your team learns to build and maintain models on the same tools, and we review the first ones with them.

3

You run the tools

License the software and keep the model current as the plant changes. We stay available as extra modeling capacity.

Validation

Why our models hold up

A model earns the right to predict by matching history first. In a peer-reviewed study at the 2020 Winter Simulation Conference, a reliability model of a food plant matched a full year of measured OEE to within one percentage point, failure mode by failure mode.

That model was later rebuilt in ReliaSim and independently validated to the same standard. Read the published validation case study.

Tell us about the decision

Describe the line, plant or network and what you need to decide. We'll tell you whether simulation is the right tool and what a validated answer would take.