How It Works

From intro call to final report — what happens when you work with Margaux Systems.

How it works

Every engagement follows the same predictable path.

1 30 min

Intro call

We discuss your operational challenge, what you've tried, and what you need to decide. No preparation needed — just the question you're trying to answer.

2 ~1 week

Data collection

You share relevant data — production routings, demand patterns, machine capacities, scheduling constraints, and cost assumptions. We review, clarify, and identify what's needed vs. what's available.

3 1-2 weeks

Model & analyze

We build a constraint-based or simulation model of your system, run alternative scenarios, and quantify trade-offs. The model captures your actual constraints, not idealized assumptions.

4 1 day

Report & discuss

You receive a written report with findings, quantified scenario comparisons, and a clear recommendation. We walk through it together so you understand not just what the answer is, but why.

No long-term commitment. Each engagement is scoped upfront with a fixed price.

What we model

We use three approaches, chosen based on the problem:

  • Operations Research — optimization models for planning, resource allocation, and flow. Best when you need to find the best configuration among many possibilities.
  • Constraint Programming — models for complex scheduling with interdependent rules. Best when feasibility is hard and the solution space is combinatorial.
  • Simulation — stochastic models for uncertainty, variability, and “what-if” exploration. Best when randomness matters (arrival patterns, machine downtime, demand volatility).

Most engagements combine at least two.


What you need to provide

The input data varies by problem, but typically includes:

  • Production routings — which operations happen on which machines, in what order
  • Processing times — how long each operation takes (deterministic or distributions)
  • Demand — what needs to be produced, by when
  • Capacity — machine availability, shifts, maintenance windows
  • Cost assumptions — tardiness penalties, labor rates, CAPEX budgets

If you don’t have everything, we work with what’s available and flag assumptions clearly.


What you receive

Every engagement ends with a written report containing:

  • The question — restated with context, so it’s clear what was asked
  • The model — what was modeled, key assumptions, and why they were chosen
  • The scenarios — each alternative tested, with quantified outcomes
  • The comparison — side-by-side KPIs (throughput, tardiness, cost, utilization) across scenarios
  • The recommendation — which option we’d recommend, with expected impact and ROI

Reports are designed to be shared with stakeholders who weren’t in the room.


Example output

The same job shop, scheduled two different ways. Drag the slider to compare.

Same jobs, same machines. Drag the slider to see the difference.

One schedule respects real constraints. The other shows what’s possible with the right intervention. This is the kind of comparison every report includes.

Ready to start?

Tell us what you're trying to decide. We'll suggest an approach and scope — no commitment.

Start the conversation →