System and method of discrete planning for process industry

1. A system of supply chain planning of meat process industry production, comprising: a computer, comprising a processor and memory, the computer configured to:

model a supply chain planning problem for two or more products of a meat process industry, wherein a coproduct is produced for at least one of the products;

group the two or more products into groups, wherein the products in each of the groups share a same type of a first dimension;

receive a weight and a yield for each raw material that produces each of the products in at least one of the groups; cluster each of the raw materials using weight-yield clustering;

generate bill of material grouping based, at least in part, on the clustered raw materials;

assign one bill of material grouping to each of the raw materials of a single cluster, wherein at least one finished good is produced from any of the raw materials assigned to the one bill of material grouping, and wherein the one bill of material grouping reduces a quantity of produced co-products, which reduces an amount of processed product that is wasted; and

generate instructions based, at least in part, on the assigned one bill of material grouping, wherein the generated instructions instruct one or more automated machines to produce processed goods comprising one or more special cuts for a meat processor, and further wherein the generated instructions reduce an amount of meat that is wasted or placed in frozen storage.

2. The system of claim 1 , wherein the computer is further configured to:

group the products sharing the same type of the first dimension;

group the products in each of the first dimension groups according to a second dimension; and

calculate a demand quantity of a material in a supply chain, wherein the coproduct produced for the at least one of the products satisfies a demand for at least one of the finished goods.

3. The system of claim 1 , wherein the computer is configured to cluster each of the raw materials using weight-yield clustering by:

plotting the yield versus the weight for each of the raw materials in the at least one of the groups as a weight-yield plot;

selecting an initial quantity of centroids for the weight-yield plots for the raw materials in the at least one of the groups;

iteratively calculating a Euclidean distance from each weight-yield plot to each of the centroids;

clustering each of the raw materials to a centroid of having the minimum Euclidean distance; and

calculating a new centroid for each of the clusters until the difference between the two consecutive iteratively-calculated means is less than a predetermined tolerance value.

4. The system of claim 1 , wherein the yield is the ratio of co-products determined by a reverse bill of material.

5. The system of claim 1 , wherein the computer is further configured to:

solve the supply chain planning problem using linear programming (LP) optimization.

6. The system of claim 2 , wherein the supply chain is a meat processing supply chain, the first dimension is a part type and the second dimension is a cut type.

7. The system of claim 5 , wherein the supply chain planning problem comprises at least one objective function, the at least one objective function is a WIP flush objective.

8. A method of supply chain planning of meat process industry production, comprising:

modeling, by a computer comprising a processor and memory, a supply chain planning problem for two or more products of a meat process industry, wherein a coproduct is produced for at least one of the products;

grouping, by the computer, the two or more products into groups, wherein the products in each of the groups share a same type of a first dimension;

receiving, by the computer, a weight and a yield for each raw material that produces each of the products in at least one of the groups;

clustering, by the computer, each of the raw materials using weight-yield clustering;

generating, by the computer, bill of material grouping based, at least in part, on the clustered raw materials;

assigning, by the computer, one bill of material grouping to each of the raw materials of a single cluster, wherein at least one finished good is produced from any of the raw materials assigned to the one bill of material grouping, and wherein the one bill of material grouping reduces a quantity of produced co-products, which reduces an amount of processed product that is wasted; and

generating, by the computer, instructions based, at least in part, on the assigned one bill of material grouping, wherein the generated instructions instruct one or more automated machines to produce processed goods comprising one or more special cuts for a meat processor, and further wherein the generated instructions reduce an amount of meat that is wasted or placed in frozen storage.

9. The method of claim 8 , further comprising:

grouping, by the computer, the products sharing the same type of the first dimension;

grouping, by the computer, the products in each of the first dimension groups according to a second dimension; and

calculating, by the computer, a demand quantity of a material in a supply chain, wherein the coproduct produced for the at least one of the products satisfies a demand for at least one of the finished goods.

10. The method of claim 8 , wherein clustering each of the raw materials using weight-yield clustering comprises:

plotting the yield versus the weight for each of the raw materials in the at least one of the groups as a weight-yield plot;

selecting an initial quantity of centroids for the weight-yield plots for the raw materials in the at least one of the groups;

iteratively calculating a Euclidean distance from each weight-yield plot to each of the centroids;

clustering each of the raw materials to a centroid of having the minimum Euclidean distance; and

calculating a new centroid for each of the clusters until the difference between the two consecutive iteratively-calculated means is less than a predetermined tolerance value.

11. The method of claim 8 , wherein the yield is the ratio of co-products determined by a reverse bill of material.

12. The method of claim 8 , further comprising:

solving, by the computer, the supply chain planning problem using linear programming (LP) optimization.

13. The method of claim 9 , wherein the supply chain is a meat processing supply chain, the first dimension is a part type and the second dimension is a cut type.

14. The method of claim 12 , wherein the supply chain planning problem comprises at least one objective function, the at least one objective function is a WIP flush objective.

15. A non-transitory computer-readable medium embodied with software, the software when executed:

models a supply chain planning problem for two or more products of a meat process industry, wherein a coproduct is produced for at least one of the products;

groups the two or more products into groups, wherein the products in each of the groups share a same type of a first dimension;

receives a weight and a yield for each raw material that produces each of the products in at least one of the groups;

clusters each of the raw materials using weight-yield clustering;

generates bill of material grouping based, at least in part, on the clustered raw materials;

assigns one bill of material grouping to each of the raw materials of a single cluster, wherein at least one finished good is produced from any of the raw materials assigned to the one bill of material grouping, and wherein the one bill of material grouping reduces a quantity of produced co-products, which reduces an amount of processed product that is wasted; and

generates instructions based, at least in part, on the assigned one bill of material grouping, wherein the generated instructions instruct one or more automated machines to produce processed goods comprising one or more special cuts for a meat processor, and further wherein the generated instructions reduce an amount of meat that is wasted or placed in frozen storage.

16. The non-transitory computer-readable medium of claim 15 , wherein the software when executed further:

groups the products sharing the same type of the first dimension;

groups the products in each of the first dimension groups according to a second dimension; and

calculates a demand quantity of a material in a supply chain, wherein the coproduct produced for the at least one of the products satisfies a demand for at least one of the finished goods.

17. The non-transitory computer-readable medium of claim 15 , wherein clusters each of the raw materials using weight-yield clustering comprises:

plotting the yield versus the weight for each of the raw materials in the at least one of the groups as a weight-yield plot;

selecting an initial quantity of centroids for the weight-yield plots for the raw materials in the at least one of the groups;

iteratively calculating a Euclidean distance from each weight-yield plot to each of the centroids;

clustering each of the raw materials to a centroid of having the minimum Euclidean distance; and

calculating a new centroid for each of the clusters until the difference between the two consecutive iteratively-calculated means is less than a predetermined tolerance value.

18. The non-transitory computer-readable medium of claim 15 , wherein the yield is the ratio of co-products determined by a reverse bill of material.

19. The non-transitory computer-readable medium of claim 15 , wherein the software when executed further:

solves the supply chain planning problem using linear programming (LP) optimization.

20. The non-transitory computer-readable medium of claim 19 , wherein the supply chain planning problem comprises at least one objective function, the at least one objective function is a WIP flush objective.