Supply chain consulting
Supply chain decisions, backed by math you can check.
We build the network models, simulations, and forecasts behind decisions that are expensive to reverse — where to put a facility, how much buffer to hold, what a route change actually costs. Then we hand you the model and show your team how to run it.
What we are
A small analytics practice, not a staffing firm.
Most supply chain questions worth paying for are quantitative, and most of them are answered with a slide deck. We do the other thing: build the model, test it against reality, and let the numbers settle the argument.
Engagements are run by the people who do the modeling. There is no pyramid, no rotating bench, and no phase two proposal attached to the final readout. The work ends when your team can rerun it without us.
Services
Seven practice areas, one modeling discipline.
Most engagements draw on two or three of these. They share a data layer and a standard of proof, which is why combining them does not multiply the timeline.
Supply Chain Network Design
Where things should be made, stored, and shipped from.
ExploreSupply Chain Simulation and Digital Twins
Test the change before it touches the floor.
ExploreDemand Forecasting and Planning
Forecasts that planners trust enough to act on.
ExploreInventory Optimization
Less working capital, better service, on purpose.
ExploreTransportation and Route Optimization
Fewer miles, fuller trailers, tighter windows.
ExploreS&OP and Integrated Business Planning
One plan the whole business is actually holding.
ExploreSupply Chain Analytics and Data Engineering
The plumbing that keeps the models honest.
ExploreNot sure which of these your problem is? That is a normal place to start. Describe the decision and we will tell you what it would take to answer it.
Talk it through
How we work
Three commitments that shape every engagement.
These are the parts clients tell us are different. They also explain why our timelines have a data reconciliation phase that other proposals do not.
The baseline ties out first
Before a single scenario runs, the model reproduces last year of actual cost by category, reconciled with finance. Studies that skip this step get unwound in the readout by one good question.
You keep the model
Deliverables are working models with documentation and a trained team, not a deck describing a model we take home with us. If you cannot rerun it in six months, we did not finish.
Ranges, not false precision
Every number arrives with the assumption it depends on and the range it moves across. A single confident figure with no sensitivity behind it is a marketing claim, not an analysis.
Engagement shape
A typical engagement runs in four phases.
Timelines vary by scope, but the sequence does not. Nothing gets analyzed before the baseline reconciles, and nothing closes before the handover is done.
- 011 to 2 weeks
Scope
We agree on the decision the work has to inform, what would change the answer, and what data exists. Scope is written down as questions, not as workstreams.
- 022 to 4 weeks
Baseline
Data reconciliation and a calibrated model of current state. This is where surprises surface, and it is deliberately early rather than convenient.
- 033 to 6 weeks
Analyze
Optimization, simulation, and scenario work against the agreed questions, reviewed in working sessions rather than saved for a reveal.
- 041 to 2 weeks
Hand over
Model handover, documentation, training, and the operating cadence that keeps the parameters current after we leave.
Industries
Where the constraints differ, so does the model.
The math generalizes. The constraints do not. Shelf life, retailer compliance windows, driver hours, and validation requirements each change the shape of the problem.
Consumer Packaged Goods
High volume, thin margins, and retail customers whose compliance programs turn a late truck into a chargeback.
Retail and E-Commerce
Omnichannel fulfillment where the same unit can be picked, shipped, or sold in store, and the cheapest choice changes hourly.
Industrial Manufacturing
Deep bills of material, long lead time components, and a plant schedule that is expensive to disturb.
Life Sciences and Pharma
Cold chain integrity, serialization, and validated processes where the cost of a stockout is not measured in dollars alone.
Food and Beverage
Perishability, seasonality, and a delivery network that runs every day regardless of what the plan said.
Distribution and 3PL
Service commitments made to other companies, on assets you have to keep utilized to make the economics work.
Representative work
What these engagements look like in practice.
- Network DesignConsumer Packaged Goods
Consolidating a nine site distribution footprint
A network that had grown by acquisition was carrying duplicate coverage in three regions. Optimization showed the footprint could contract without losing next day reach.
6Sites in recommended footprint - SimulationRetail and E-Commerce
Sizing an automation investment before signing
A vendor proposal promised throughput that the demand profile never actually required, and understated the staffing needed on the days that mattered.
SmallerConfiguration change - ForecastingIndustrial Manufacturing
Forecasting a long tail service parts catalog
Exponential smoothing was being applied uniformly to a catalog where most items moved a handful of times a year, producing confident forecasts of zero.
100%Items on a fitted method
Illustrative composites based on common engagement patterns. Figures show the type of outcome these methods target and are not audited results or a promise of comparable performance.
Insights
Writing from the practice.
Notes on modeling, planning, and the failure modes we keep running into.
- Modeling6 min read
Optimization and simulation answer different questions
Teams often buy one and expect the other. The distinction is not academic; it determines whether your plan survives a bad week.
- Forecasting7 min read
Forecast accuracy is the wrong target
A forecast that is two points more accurate and changes no decision has produced nothing. Measure the cost of the errors, not the size of them.
- Network Design5 min read
If the baseline does not tie out, nothing downstream matters
The least glamorous phase of a network study is the one that determines whether anyone believes the result.
Next step
Tell us the decision you are trying to make.
A first conversation is a working session, not a pitch. We will tell you what it would take to answer the question properly, and whether it is worth answering at all.