MilkBot describes a lactation in terms of its scale, ramp, and persistence. These parameters control the shape of the lactation curve.
The fitting engine uses breed, parity, and any test day data points to calculate values for each parameter. These describe the lactation as a whole, and can be used to calculate expected milk at each day of the lactation as well as M305 and peak milk.
Starting with no data points, MilkBot's best guess will be the breed-parity average, but with each added point it can make an improved summary of a particular lactation