Data-driven approach to modeling sensors wherein optimal time delays are determined for a first set of predictors and stored as a second set of predictors

A method, computer program product and system are provided for modeling non-controllable parameters affecting system performance. The method may include receiving historical values for each of a plurality of system parameters and grouping the system parameters into controllable, non-controllable, and performance parameters. The method may further include determining a first set of predictors from the non-controllable parameters using the historical values of these non-controllable parameters and, for each predictor in the first set, determining optimal time instances at which a value of each predictor is measured using non-uniform time scales. These optimal time instances may then be saved as a second set of predictors. One or more constraints may then be established for each of the controllable parameters. Finally, a dynamic model based on the second set of predictors, the controllable parameters, and the performance parameters may be constructed and optimized.