Simulation consulting
How a simulation project runs
Four steps, from the decision in front of you to a model your team keeps using.
From the question to a tested answer
The model is scoped to one decision and checked against how the operation really ran.
Name the decision
What are you about to spend, change or commit to? The model is scoped to that decision and leaves the rest of the building out.
Build from your data
Rates, failures, repair times, schedules, demand and forecasts come from your own records. The model is checked against how the operation really ran.
Run the experiments
We vary the factors that matter, such as buffer size, run length or staffing, and show where the returns start to flatten.
Keep it working
Many clients keep using the model. Any model we make can become an app your planners run themselves, with only a little more effort.
Sometimes the model is done before it runs
A client planned a new operation and asked us to model it. As we worked through how the model would handle a machine running out of supplies or breaking down, it became clear the system could not be run as designed. The operation leaders called and cancelled all of their long-lead equipment orders.
A simulation has to march forward through time and deal with every realistic event, so the specification cannot be fudged.
From one machine to a whole network
Some questions live on the plant floor. Others live in the boardroom. The same model can often serve both.
On the plant floor
- Line modeling: from a single operation to a complete manufacturing system
- Multi-stage plants: processes that depend on each other
- Capacity planning: what the system can really do, and what limits it
- Bottleneck analysis: find the constraint and test the fix
In the budget meeting
- Investment justification: put numbers on the benefit of a capital project
- Scenario evaluation: test an operating change with no risk to production
- Long-term planning: growth options, compared side by side
- Supply chain networks: trade-offs across facilities, end to end
Under the hood
- Discrete event simulation with statistical analysis of the results
- Discrete rate simulation for high-speed and bulk flow
- Reliability modeling of stops, repairs and maintenance plans
- Integration with the manufacturing systems you already run
Rate-based models for high-speed lines
Andy Siprelle created bulk flow simulation in 1990 to model lines that move product as a flow rather than one part at a time. It later became discrete rate simulation. We still use it on packaging, food and bulk-handling lines where a conventional model would be slow to run.
How discrete rate simulation worksWhen the model should become an app
A model that answers one question well often gets asked the next one. We wrap models in apps that planners, analysts and finance teams run on their own, in the language of their job.
One started as a six-hour model and became a 20-minute tool a CFO used for the annual budget.
Custom applicationsTell us your problem
What would happen if you changed it?
Describe the decision in front of you. We'll tell you whether a model can answer it, and what it would take.
- A straight answer on whether simulation is the right tool
- Which tool fits, even if it isn't one we use every day
- A rough scope and timeline, before any commitment