System and method for automatic parameter tuning of campaign planning with hierarchical linear programming objectives
A system and method are disclosed for campaign planning and include modeling the use of campaign operations and campaignable resources of a supply chain network including a production line to produce products using campaign operations and campaignable resources as campaign planning problems, defining an evaluation function comprising a weighted sum of features evaluated from the campaign planning problem, initializing weights to build a consumption profile and evaluation function, determining fitness values that indicate a level of variability, evaluating reward values based on the fitness values, selecting a sub-sample of the top fitness values having the best associated objective function, repeating the generating, the evaluating and the selecting steps to adjust the weights until a stopping criterion is met indicating an optimal solution has been reached, and determining a campaign plan for the use of the campaign operations and campaignable resource.
1. A system of campaign planning, comprising:
a computer, comprising a processor and a memory, configured to:
model a use of one or more campaign operation and one or more campaignable resource of a supply chain network, wherein the one or more campaignable resource are utilized in a batch or a continuous process;
define an evaluation function comprising a weighted sum of features evaluated from a campaign planning problem;
initialize weights to build a consumption profile and evaluation function;
determine fitness values, wherein the fitness values indicate a level of variability;
evaluate reward values based on the fitness values;
select a sub-sample of top fitness values having a best associated objective function;
repeat the determine, the evaluate and the select to adjust the weights until a stopping criterion is met indicating an optimal solution has been reached; and
determine a campaign plan for a use of the one or more campaign operation and the one or more campaignable resource.
2. The system of claim 1 , wherein the weights are adjusted using a cross-entropy method.
3. The system of claim 1 , wherein the supply chain network further comprises operations, buffers and pathways.
4. The system of claim 1 , wherein the modelling further comprises encoding a required policy of a sequential decision problem into a k-lookahead search strategy.
5. The system of claim 1 , wherein the campaign plan further comprises an inventory policy.
6. The system of claim 5 , wherein the inventory policy comprises a reorder point and a target quantity.
7. A computer-implemented method of campaign planning, comprising:
modeling, by a computer, a use of one or more campaign operation and one or more campaignable resource of a supply chain network, wherein the one or more campaignable resource are utilized in a batch or a continuous process;
defining an evaluation function comprising a weighted sum of features evaluated from a campaign planning problem;
initializing weights to build a consumption profile and evaluation function;
determining fitness values, wherein the fitness values indicate a level of variability;
evaluating reward values based on the fitness values;
selecting a sub-sample of top fitness values having a best associated objective function;
repeating the determining, the evaluating and the selecting to adjust the weights until a stopping criterion is met indicating an optimal solution has been reached; and
determining, by the computer, a campaign plan for a use of the one or more campaign operation and the one or more campaignable resource.
8. The computer-implemented method of claim 7 , wherein the weights are adjusted using a cross-entropy method.
9. The computer-implemented method of claim 7 , wherein the supply chain network further comprises operations, buffers and pathways.
10. The computer-implemented method of claim 7 , wherein the modelling further comprises encoding a required policy of a sequential decision problem into a k-lookahead search strategy.
11. The computer-implemented method of claim 7 , wherein the campaign plan further comprises an inventory policy.
12. The computer-implemented method of claim 11 , wherein the inventory policy comprises a reorder point and a target quantity.
13. A non-transitory computer-readable medium embodied with software for campaign planning, the software when executed:
models a use of one or more campaign operation and one or more campaignable resource of a supply chain network, wherein the one or more campaignable resource are utilized in a batch or a continuous process;
defines an evaluation function comprising a weighted sum of features evaluated from a campaign planning problem;
initializes weights to build a consumption profile and evaluation function;
determines fitness values, wherein the fitness values indicate a level of variability;
evaluates reward values based on the fitness values;
selects a sub-sample of top fitness values having a best associated objective function;
repeats the determines, the evaluates and the selects to adjust the weights until a stopping criterion is met indicating an optimal solution has been reached; and
determines a campaign plan for a use of the one or more campaign operation and one or more campaignable resource.
14. The non-transitory computer-readable medium of claim 13 , wherein the weights are adjusted using a cross-entropy method.
15. The non-transitory computer-readable medium of claim 13 , wherein the supply chain network further comprises operations, buffers and pathways.
16. The non-transitory computer-readable medium of claim 13 , wherein the modelling further comprises encoding a required policy of a sequential decision problem into a k-lookahead search strategy.
17. The non-transitory computer-readable medium of claim 13 , wherein the campaign plan further comprises an inventory policy comprising a reorder point and a target quantity.