If I may put a different spin on Old Smokey's comments.
(a) the AFM data generally originates from various mathematical models developed from first principles (although, in practice, one starts with any of the "standard" sets of equations). The windtunnel and flight test programs are used to validate these models .. which often involves the need for adopting fudge factor fixes to make the equations fit the observations.
Once an acceptable model is developed (and accepted by the regulatory authority) that model is used to build whatever AFM charts might be desired.
(b) if an operator chooses not to use the manufacturer's models for some reason, then the operator can go through the exercise of setting up extensive lookup tables or doing the regressions to derive a large set of equations to represent the AFM carpets. The result of either is a workable PC system to punch out RTOW data. Doing the work manually .. ie direct from the AFM .. is VERY labour intensive and only really suitable for a small aerodrome network