Hopefully there's good empirical data powering the model here, which just added slump prediction:

> Alongside the event, Meta is releasing a new AI model for designing concrete mixes, Bayesian Optimization for Concrete (BOxCrete). BOxCrete improves over Meta’s previous models with more robustness to noisy data as well as new features including the ability to predict concrete slump (an important indicator of concrete workability).

Seems hard to imagine not doing a slump test, trusting AI when it comes to your multi/many million dollar build outs for something so important. But perhaps still useful for planning, as a starting place?

There is a lot of expense in the wasted concrete from all the different pours that are slump tested. There is a lot of cost from concrete that leaves the plant only to fail slump testing when it gets there - not only do you have to empty the truck someplace else, but there maybe contract provisions if you fail to keep the workers busy. Often more than one test is done - if the plant has an order that they know will be tested they test before it leaves the plant (if it fails they can redirect to a different customer who will knowingly accept lesser concrete - but concrete cures on the truck so it would be unacceptable if rejected at the site to go elsewhere). Many smaller jobs skip the test, but they would like it done if it was free.

That said, I'm not sure if the value can ever be greater than a slump test just before pouring.

The predictions of the model are used as recommendations for onsite testing to accelerate finding mixtures with optimal strength-speed-sustainability trade-offs. We are not replacing canonical testing with the model.