Predictive process control for a manufacturing process
Aspects of the disclosed technology encompass the use of a deep-learning controller for monitoring and improving a manufacturing process. In some aspects, a method of the disclosed technology includes steps for: receiving control values associated with a process station in a manufacturing process, predicting an expected value for an article of manufacture output from the process station, and determining if the deep-learning controller can control the manufacturing process based on the expected value. Systems and computer-readable media are also provided.
1. A computer-implemented method for training a deep-learning controller to control a manufacturing process executed in a manufacturing system comprising a plurality of process stations, comprising:
receiving, by the deep-learning controller, a first set of control values associated with a first process station of the plurality of process stations, wherein the first set of control values are attributes of the first process station;
predicting, by the deep-learning controller, an expected final output for an article of manufacture, based on the first set of control values;
receiving, by the deep-learning controller, a second set of control values associated with a second process station of the plurality of process stations, wherein the second set of control values are attributes of the second process station;
predicting, by the deep-learning controller, an updated expected final output for the article of manufacture, based on the second set of control values;
determining if the deep-learning controller can control the manufacturing process based on the expected final output and the updated expected final output by:
receiving, by the deep-learning controller, an actual value associated with the article of manufacture following processing in the manufacturing system,
comparing the expected final output with the actual value, and
determining that the expected final output is within a threshold confidence level of the actual value,
comparing the updated expected final output with the actual value, and
determining that the updated expected final output is within the threshold confidence level of the actual value; and
based on the determining, assuming control of the manufacturing process by the deep-learning controller.
2. The computer-implemented method of claim 1 , further comprising:
indicating that the deep-learning controller is ready to take control of the manufacturing process.
3. The computer-implemented method of claim 1 , wherein
if the expected final output is not within the threshold confidence level of the actual value, determining that the deep-learning controller cannot control the manufacturing process.
4. The computer-implemented method of claim 1 , wherein the expected final output is based on two or more physical attributes of the article of manufacture.
5. A system for training a deep-learning controller to control a manufacturing process executed in a manufacturing system comprising a plurality of process stations comprising:
one or more processors; and
a non-transitory memory storing instructions that, when executed by the one or more processors, cause the one or more processors to perform operations comprising:
receiving, by the deep-learning controller, a first set of control values associated with a first process station of the plurality of process stations, wherein the first set of control values are attributes of the first process station;
predicting, by the deep-learning controller, an expected final output for an article of manufacture, based on the first set of control values;
receiving, by the deep-learning controller, a second set of control values associated with a second process station of the plurality of process stations, wherein the second set of control values are attributes of the second process station;
predicting, by the deep-learning controller, an updated expected final output for the article of manufacture, based on the second set of control values;
determining if the deep-learning controller can control the manufacturing process based on the expected final output and the updated expected final output by:
receiving, by the deep-learning controller, an actual value associated with the article of manufacture following processing in the manufacturing system,
comparing the expected final output with the actual value, and
determining that the expected final output is within a threshold confidence level of the actual value,
comparing the updated expected final output with the actual value, and
determining that the updated expected final output is within the threshold confidence level of the actual value; and
based on the determining, assuming control of the manufacturing process by the deep-learning controller.
6. The system of claim 5 , wherein the one or more processors are further configured to perform operations comprising:
indicating that the deep-learning controller is ready to take control of the manufacturing process.
7. The system of claim 5 , wherein
if the expected final output is not within the threshold confidence level of the actual value, determining that the deep-learning controller cannot control the manufacturing process.
8. The system of claim 5 , wherein the first set of control values are received from one or more process controllers associated with the first process station.
9. The system of claim 5 , wherein the expected final output is based on two or more physical attributes of the article of manufacture.
10. A non-transitory computer-readable storage medium for training a deep-learning controller to control a manufacturing process executed in a manufacturing system comprising a plurality of process stations, the non-transitory computer-readable storage medium comprising instructions stored therein, which when executed by one or more processors, cause the one or more processors to perform operations comprising:
receiving, by the deep-learning controller, a first set of control values associated with a first process station of the plurality of process stations, wherein the first set of control values are attributes of the first process station;
predicting, by the deep-learning controller, an expected final output for an article of manufacture, based on the first set of control values;
receiving, by the deep-learning controller, a second set of control values associated with a second process station of the plurality of process stations, wherein the second set of control values are attributes of the second process station;
predicting, by the deep-learning controller, an updated expected final output for the article of manufacture, based on the second set of control values;
determining if the deep-learning controller can control the manufacturing process based on the expected final output and the updated expected final output by:
receiving, by the deep-learning controller, an actual value associated with the article of manufacture following processing in the manufacturing system,
comparing the expected final output with the actual value, and
determining that the expected final output is within a threshold confidence level of the actual value,
comparing the updated expected final output with the actual value, and
determining that the updated expected final output is within the threshold confidence level of the actual value; and
based on the determining, assuming control of the manufacturing process by the deep-learning controller.
11. The non-transitory computer-readable storage medium of claim 10 , wherein the one or more processors are further configured to perform operations comprising:
indicating that the deep-learning controller is ready to take control of the manufacturing process.
12. The non-transitory computer-readable storage medium of claim 10 , wherein
if the expected final output is not within the threshold confidence level of the actual value, determining that the deep-learning controller cannot control the manufacturing process.
13. The non-transitory computer-readable storage medium of claim 10 , wherein the first set of control values are received from one or more process controllers associated with the first process station.