Graph theory and network analytics and diagnostics for process optimization in manufacturing
A system, method, and computer-readable medium are disclosed for analysis and characterization of manufacturing information such as process trees or genealogies using graph theory. More specifically, using graph theory to analyze manufacturing information of a manufacturing operation allows for deep analysis of relationships between batches or units in a process tree and their closeness or distance, to identify clusters associated with specific quality characteristics or problems, to identify common antecedents of specifically labeled batches (e.g., problem batches), and/or to detect overall desirable or undesirable characteristics of the process tree (e.g., centrality, etc.).
1. A computer-implementable method for process optimization by a special purpose computer in a complex manufacturing operation, comprising:
providing a memory comprising an operating system and a graph theory manufacturing operation representation portion;
identifying, by a special purpose computer using instructions comprised in the memory, manufacturing units of a manufacturing operation, the manufacturing units comprising manufacturing components and manufacturing operations;
characterizing, by the special purpose computer using the instructions comprised in the memory, the manufacturing units as nodes for inclusion on a manufacturing operation graph representation;
characterizing, by the special purpose computer using the instructions comprised in the memory, at least one of input materials, suppliers, parts, and other inputs into the manufacturing operation as nodes for inclusion on the manufacturing operation graph representation;
characterizing, by the special purpose computer using the instructions comprised in the memory, manufacturing steps within the manufacturing operation as connections for inclusion on the manufacturing operation graph representation;
generating, by the special purpose computer using the instructions comprised in the memory, the manufacturing operation graph representation using the characterized nodes and connections;
measuring, by the special purpose computer using the instructions comprised in the memory, a quality characteristic of a batch associated with the manufacturing operation;
identifying, by the special purpose computer using the instructions comprised in the memory, connections between the measured batch and an unmeasured batch associated with the manufacturing operation;
computing, by the special purpose computer using the instructions comprised in the memory, and based on the manufacturing operation graph representation and the identified connections, a degree of connectedness between the measured batch and the unmeasured batch; and
predicting, by the special purpose computer using the instructions comprised in the memory and based on the degree of connectedness, a quality characteristic of the unmeasured batch.
2. The method of claim 1 , further comprising:
performing graph analytics on the manufacturing operation graph representation.
3. The method of claim 2 , wherein the measured batch comprises known bad batch.
4. The method of claim 2 , further comprising:
clustering, in a first cluster and by the special purpose computer using the instructions comprised in the memory, first batches associated with the manufacturing operation;
clustering, in a second cluster and by the special purpose computer using the instructions comprised in the memory, second batches associated with the manufacturing operation;
determining, by the special purpose computer using the instructions comprised in the memory, differences between the first cluster and the second cluster, based on common antecedent batches between the first cluster and the second cluster.
5. The method of claim 4 , wherein the connections comprise at least one uni-directional connection, the uni-directional connection defining a workflow of the first batches.
6. The method of claim 5 , wherein the connections comprise at least one bi-directional connection, the bi-directional connection defining a workflow of the second batches.
7. The method of claim 6 , wherein the connections identify relationships of the manufacturing items through upstream nodes and downstream nodes.
8. The method of claim 7 , wherein in response to determining the quality characteristic of the unmeasured batch does not meet a quality standard, the unmeasured batch is reused as an input material.
9. The method of claim 8 , wherein the graph analytics are used to determine a centrality characteristic of the manufacturing operation graph representation.
10. The method of claim 8 , wherein the manufacturing operation comprises a pharmaceutical manufacturing operation.
11. The method of claim 8 , wherein the manufacturing operation comprises an information handling system manufacturing operation.
12. A special purpose computer system for process optimization in a complex manufacturing operation, the special purpose computer system comprising:
a processor;
a data bus coupled to the processor; and
a non-transitory, computer-readable storage medium embodying computer program code, the non-transitory, computer-readable storage medium being coupled to the data bus, the computer-readable storage medium comprising an operating system and a graph theory manufacturing operation representation portion, the computer program code interacting with a plurality of computer operations and comprising instructions executable by the processor and configured for:
identifying manufacturing units of a manufacturing operation;
characterizing the manufacturing units as nodes for inclusion on a manufacturing operation graph representation;
characterizing at least one of input materials, suppliers, parts, and other inputs into the manufacturing operation as nodes for inclusion on the manufacturing operation graph representation;
characterizing manufacturing steps within the manufacturing operation as connections for inclusion on the manufacturing operation graph representation;
generating the manufacturing operation graph representation using the characterized nodes and connections;
measuring a quality characteristic of a batch associated with the manufacturing operation;
identifying connections between the measured batch and an unmeasured batch associated with the manufacturing operation;
computing, using the manufacturing operation graph representation, and based on the identified connections, a degree of connectedness between the measured batch and the unmeasured batch; and
predicting, based on the degree of connectedness, a quality characteristic of the unmeasured batch.
13. The system of claim 12 , wherein the instructions executable by the processor are further configured for:
performing graph analytics on the manufacturing operation graph representation.
14. The system of claim 12 , wherein the measured batch comprises a known bad batch.
15. The system of claim 12 , wherein the instructions executable by the processor are further configured for:
clustering, in a first cluster and by the special purpose computer using the instructions comprised in the memory, first batches associated with the manufacturing operation;
clustering, in a second cluster and by the special purpose computer using the instructions comprised in the memory, second batches associated with the manufacturing operation;
determining, by the special purpose computer using the instructions comprised in the memory, differences between the first cluster and the second cluster, based on common antecedent batches between the first cluster and the second cluster.
16. A non-transitory, computer-readable storage medium embodying computer program code for achieving process optimization in a complex manufacturing operation, the computer-readable storage medium comprising a graph theory manufacturing operation representation portion, the computer program code comprising computer executable instructions configured for:
identifying manufacturing units of a manufacturing operation;
characterizing the manufacturing units as nodes for inclusion on a manufacturing operation graph representation;
characterizing at least one of input materials, suppliers, parts, and other inputs into the manufacturing operation as nodes for inclusion on the manufacturing operation graph representation;
characterizing manufacturing steps within the manufacturing operation as connections for inclusion on the manufacturing operation graph representation;
generating the manufacturing operation graph representation using the characterized nodes and connections;
measuring a quality characteristic of a batch associated with the manufacturing operation;
identifying connections between the measured batch and an unmeasured batch associated with the manufacturing operation;
computing, using the manufacturing operation graph representation, and based on the identified connections, a degree of connectedness between the measured batch and the unmeasured batch; and
predicting, based on the degree of connectedness, a quality characteristic of the unmeasured batch.
17. The non-transitory, computer-readable storage medium of claim 16 , wherein the computer executable instructions are further configured for:
performing graph analytics on the manufacturing operation graph representation.
18. The non-transitory, computer-readable storage medium of claim 16 , wherein the measured batch comprises a known bad batch.
19. The non-transitory, computer-readable storage medium of claim 16 , wherein the computer executable instructions are further configured for:
clustering, in a first workflow and by the special purpose computer using the instructions comprised in the memory, first batches associated with the manufacturing operation;
clustering, in a second workflow and by the special purpose computer using the instructions comprised in the memory, second batches associated with the manufacturing operation;
determining, by the special purpose computer using the instructions comprised in the memory, differences between the first cluster and the second cluster, based on common antecedent batches between the first cluster and the second cluster.