I build tools that make good decisions under uncertainty. Mostly logistics.
A synthetic city, not real data.
Most decisions that matter are made before the facts are in. I build for that moment.
How many loaves to bake before you know the demand.1 Which order to drive before you know how heavy the day will be.2 The usual answer is a forecast and a plan built on it.
I am more interested in a plan that still holds when the forecast is wrong, and in showing honestly how often that happens.
So everything here is built the same way: a real decision you can make yourself, an exact solver that makes it too, and hundreds of sampled days to see who was right.3
Drive-in racking gives a freezer 38% more places and, at 7 pallets per product, holds under a third of the pallets. Where density pays, rack type by rack type, with the numbers.
Read the essayDemos you can play and tools you can use. Each one runs in your browser, nothing is sent anywhere.
Demo
Drive an electric van's day one block at a time, then see what a weight-aware planner saves. Over 300 sampled days, the median saving is 7.6%.
Demo · In the works
Book trucks on the spot market or keep your own. The answer depends on a demand you don’t know yet.
Tool
Loading meters, pallets per trailer and the weight a carrier will bill, for a mixed consignment, drawn on the trailer floor. Seven published carrier rules and the generic convention, side by side.
Tool
From a packing list of cartons to the pallets it makes: cartons per layer, layers under the height cap, pallets to book, each one drawn with its arithmetic. Hands the pallets to Trailer fit.
Tool
How many of your pallets a cold room really holds, by rack type: selective, mobile, drive-in, push-back, flow and pallet shuttle, drawn to scale in 3D. A supplier counts places; this counts your stock on an average day.
Industrial engineering taught me to decide with numbers. Shipping software taught me the numbers have to survive a real Tuesday.