S&OP Optimization S&OP OptimizationThe optimization procedure relies on standard mathematical programming tools (namely linear and integer programming). Classically, if a resource is defined as integer, the check box Integer values can be accessed. If this cell is not selected, then the optimization process is made with real values (via linear programming and the Simplex algorithm). On the contrary, if the cell is selected, an integer optimization algorithm will be launched (namely a Branch and Bound algorithm).
Two different objective functions can be selected (Files menu, Preferences window, S&OP panel):
•the total cost over the horizon has to be minimized,
•or the total margin over the horizon has to be maximized.
Nota: in this last case, the margin is defined is the gross margin (defined in the upper level in the table) defined as the difference between the sales and the standard production costs.
The optimization procedure is launched by clicking on the Optimize button. The computation time is limited to 15 minutes (if no admissible solution is found during this time period, the process is stopped).
The first step of the optimization process consists of checking data coherence. If difficulties are detected, they are explained via messages displayed in the panel and the optimization is stopped.
At the end of the optimization process, the global results and indicators are displayed. The Optimal Value corresponds to the optimal values for the decision variables (found by the optimization procedure). Clearly, the algorithm takes into account the parameter variations introduced via the sensitivity/simulation panel.
The decision variables values can be reset to the initial values (defined in the S&OP plans by the Initial data button.
The optimized solution can be transferred into the S&OP plans (Family items and resource items) by the Transfer results button. It is worth noting that all the results will be rounded according to the required decimal number (defined in each item maintenance window).