They found a crucial intermediate step in MILP solvers that has so many possible solutions that it takes a long time to figure out, which slows down the whole process. The specialists utilized a sifting method to work on this step, then, at that point, involved AI to track down the ideal answer for a particular sort of issue.
Their information driven approach empowers an organization to utilize its own information to fit a universally useful MILP solver to the central issue.
The accuracy of MILP solvers did not suffer at all as a result of this new method’s 30 to 70% speedup. One could utilize this technique to get an ideal arrangement all the more rapidly or, for particularly complex issues, an improved arrangement in a manageable measure of time.