We are building models and policies to optimise the operations of semiconductor fabs, from optimal dispatch policies to better models of etch and lithography processes.
Decide which lot runs next on which tool as queues, set-ups and re-entrant flows change. Each policy comes with bounds on deadlines and waiting times, and is checked against them before it is used on the floor. On the Mini-Fab benchmark, we proved in Lean a lower bound over every possible policy, and found a policy that meets it exactly.
Models of etch and lithography steps, fitted to your process data, that predict a lot’s outcome before it runs.
See what a hot lot, a maintenance window or a tool going down does to cycle time before it happens.
Shift flexible energy use in time, with cycle-time commitments and process windows as hard limits.
We analyse a time-bound, anonymised set of data from your system to show the gains that Crucible can deliver.
We deploy a model on a small-scale system and measure performance.
We extend the deployment across your whole system.
Start with an assessment that quantifies the improvements Crucible can deliver.