Warehouse slotting & layout
Which SKU belongs in which location, based on real velocity and on what gets picked alongside it. Fast movers at the front and at golden-zone height; dead stock high and deep.
Optimization engineering · Monterrey, Mexico
Optimization that is computed, not estimated.
Where the fast movers sit, what order jobs run in, how many registers open at 1pm: decisions made out of habit that cost money every single day. We turn them into a mathematical model that runs on your data and hands back the plan — built one time zone away.
Interactive simulator
A plan before and after the model.
Pick a problem and watch the hand-made plan turn into the one the solver returns — with the log of the cost dropping on every iteration.
Travel distance per wave
412 m···
solving
Lines picked per hour
96···
solving
Locations to re-slot
—···
solving
Simulation with illustrative data on toy-scale scenarios. In a real project the model is formulated around your constraints — tooling, contracts, delivery windows, layout — and runs against your own history before anything changes.
Use cases
Decisions currently made by eye.
The same toolbox — integer programming, heuristics, queueing theory — pointed at each area's concrete problem.
Pick waves & order batching
Which orders get picked together, in what sequence, through which zone. A picker who walks less picks more lines per hour without moving any faster.
Production scheduling & sequencing
What runs where and in what order, so tooling families stay grouped, changeovers stop eating capacity, and promised dates still hold.
Vehicle routing & last mile
VRP and CVRP with capacity and delivery windows, solved with an engine we built and can tune — not a black box you cannot inspect when the answer looks wrong.
Queues & service capacity
How many registers, windows, or dock doors to open in each time slot. Queueing models on your own arrival history: less waiting on the same staffed hours.
Staffing & shift rostering
Cover forecast demand while respecting rest rules, skills, and contracts — with less overtime and without rebuilding the roster by hand every week.
What to expect
Industry reference figures.
These are not promises, and they are not our clients' results — they are the ranges the industry reports. The baseline measures yours before you commit.
10–30%
less picking travel
The range the industry reports after re-slotting a warehouse by real velocity instead of by product family or by habit.
$0
in new equipment
Optimization does not buy hardware: it runs on the data your ERP, WMS, or POS already stores. What changes is the decision, not the infrastructure.
Weeks
to the first real number
The model runs against your history first, so the estimated saving shows up before anything on the floor moves — and the pilot is measured against that number.
Process
How a project runs.
Step 02 is what separates this from a consulting proposal: the saving is estimated against your own history before you move a single pallet.
01
Consultation
Which decision is made by hand today, who makes it, how often, and on what data. We tell you straight whether it is worth modeling — and when it is not.
02
Baseline on your history
We formulate the problem and run the model over a past period: what it would have saved under the same conditions. Nothing on the floor changes yet.
03
Scoped pilot
One facility, one line, one site. Fixed quote and success criteria agreed before we start, measured against the baseline.
04
Integration & measurement
The model runs on its own and delivers the plan into your ERP, your WMS, or a screen on the floor. If the numbers hold, it rolls out to more areas or sites.
Same family
Other problems that model the same way.
The operation changes; the mathematics does not. If you recognize one of these as yours, it is a candidate.
Dock & yard scheduling
Appointment plans that cut carrier dwell time and the detention you end up paying for.
Load building & cube utilization
How to pack a trailer so the same volume moves in fewer trips, within weight and stacking rules.
Inventory policy
Reorder points and safety stock computed per SKU against the service level you choose.
Production & procurement mix
What to make and what to buy given margin, capacity, and the contracts already signed.
Distribution network design
Where a facility belongs and which customer it serves, when the network grew without a redesign.
Maintenance scheduling
Windows and crews assigned so planned downtime costs production as little as possible.
Where this fits
A model without data is not worth much.
Optimization is the last link: first the number has to exist, be visible, and be trusted. These usually come before it, or alongside.
Business Intelligence
If nobody today knows how far a picker walks or how long a changeover takes, the dashboard comes first — the model needs that number to have something to beat.
AI Consulting
Optimization decides; forecasting tells it what is coming. Demand prediction and anomaly detection feed the constraints the model solves against.
Custom software
For an optimized plan to reach the floor, someone has to capture real progress against it. That is the operational layer that feeds the model and receives its output.
FAQ
Frequently asked questions.
How is this different from a continuous-improvement consultancy?
Lean and Six Sigma clean up the process: fewer steps, less waste, a standard way of working. This computes the decision that somebody currently makes out of habit — which SKU goes in which location, what order jobs run in, how many registers to open at 1pm — with a model that runs on your data. They pair well: continuous improvement leaves the process tidy, and the model operates it at its best point every day.
What data do we need to start?
The history of whatever decision you want to optimize: pick lines from the WMS, work orders and cycle times from the ERP, timestamped tickets from the POS, or deliveries with time and location. It almost always already exists and nobody is using it. If the history is thin or messy, the consultation tells you that before you spend anything.
How do we know the savings are real and not a slide?
Before anything touches the floor, we run the model against your own history: we replay a past period and compare the decision that was actually made to the one the model would have produced. That number is the baseline. The pilot is then measured against it, not against an estimate.
Do we have to replace our ERP or WMS?
No. The model connects to what you already run, reads the data, and returns the plan — back into the same system, onto a dashboard, or as a file your team already knows how to use. Replacing an ERP in order to optimize is starting from the expensive end.
What if our operation changes every day?
That is exactly where this pays off most. A hand-made plan gets rebuilt weekly because rebuilding it is painful; a model reruns every morning against that day's demand, inventory, and crew. The more volatile the operation, the more the average decision costs you.
How does an engagement work?
A free consultation to scope the decision and check the data, then a pilot with a fixed quote on one site, one line, or one facility, with the success criteria agreed up front. We work from Monterrey, Mexico — the same business day and largely the same time zone as US operations, so reviews happen live instead of overnight.
What does
guessing cost?
Book a free consultation. We take one concrete decision from your operation and tell you whether it can be modeled, on what data, and what it would be worth — or whether it is not worth doing.
Talk about your operation