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 a plurality of control values from two or more stations, at a deep learning controller, wherein the control values are generated at the two or more stations deployed in a manufacturing process, predicting an expected value for an intermediate or final output of an article of manufacture, based on the control values, and determining if the predicted expected value for the article of manufacture is in-specification. In some aspects, the process can further include steps for generating control inputs if the predicted expected value for the article of manufacture is not in-specification. Systems and computer-readable media are also provided.
1. A computer-implemented method comprising:
receiving a first set of in-process inputs from a first processing station, at a deep learning controller, wherein the first set of in-process inputs are generated at the first processing station deployed in a manufacturing process, wherein the first set of in-process inputs are attributes of the first processing station of a plurality of stations in the manufacturing process;
identifying, by the deep learning controller, a final quality metric of a plurality of final quality metrics for which to optimize the manufacturing process;
generating, by the deep learning controller, an expected value of the final quality metric for an article of manufacture based on the first set of in-process inputs;
determining, by the deep learning controller, based on a comparison of the first set of in-process inputs and the expected value that anomalous activity is present; and
based on the determining, adjusting, by the deep learning controller, downstream parameters of a downstream processing station to cause the final quality metric to address the anomalous activity.
2. The computer-implemented method of claim 1 , wherein the final quality metric is a metric of a completed version of the article of manufacture that cannot be measured until the manufacturing process is complete.
3. The computer-implemented method of claim 1 , further comprising:
adjusting, by the deep learning controller, further downstream parameters of a further downstream processing station.
4. The computer-implemented method of claim 1 , further comprising:
generating, by the deep learning controller, one or more control inputs, wherein the deep learning controller adjusts the downstream parameters of the downstream processing station in accordance with the one or more control inputs.
5. The computer-implemented method of claim 1 , wherein adjusting, by the deep learning controller, the downstream parameters of the downstream processing station comprises:
identifying a key influencer associated with the anomalous activity;
identifying a control input that corresponds to the key influencer; and
adjusting the downstream parameters based on the control input corresponding to the key influencer.
6. The computer-implemented method of claim 5 , wherein the key influencer affects characteristics of outputs from the first processing station.
7. The computer-implemented method of claim 1 , further comprising:
identifying a plurality of parameters in outputs generated by the first processing station; and
determining a subset of the plurality of parameters that affect the outputs generated by the first processing station.
8. A non-transitory computer readable medium comprising one or more sequences of instructions, which, when executed by a processor, causes a computing system to perform operations comprising:
receiving a first set of in-process inputs from a first processing station, at a deep learning controller, wherein the first set of in-process inputs are generated at the first processing station deployed in a manufacturing process, wherein the first set of in-process inputs are attributes of the first processing station of a plurality of stations in the manufacturing process;
identifying, by the deep learning controller, a final quality metric of a plurality of final quality metrics for which to optimize the manufacturing process;
generating, by the deep learning controller, an expected value of the final quality metric for an article of manufacture based on the first set of in-process inputs;
determining, by the deep learning controller, based on a comparison of the first set of in-process inputs and the expected value that anomalous activity is present; and
based on the determining, adjusting, by the deep learning controller, downstream parameters of a downstream processing station to cause the final quality metric to address the anomalous activity.
9. The non-transitory computer readable medium of claim 8 , wherein the final quality metric is a metric of a completed version of the article of manufacture that cannot be measured until the manufacturing process is complete.
10. The non-transitory computer readable medium of claim 8 , further comprising:
adjusting, by the deep learning controller, further downstream parameters of a further downstream processing station.
11. The non-transitory computer readable medium of claim 8 , further comprising:
generating, by the deep learning controller, one or more control inputs, wherein the deep learning controller adjusts the downstream parameters of the downstream processing station in accordance with the one or more control inputs.
12. The non-transitory computer readable medium of claim 8 , wherein adjusting, by the deep learning controller, the downstream parameters of the downstream processing station comprises:
identifying a key influencer associated with the anomalous activity;
identifying a control input that corresponds to the key influencer; and
adjusting the downstream parameters based on the control input corresponding to the key influencer.
13. The non-transitory computer readable medium of claim 12 , wherein the key influencer affects characteristics of outputs from the first processing station.
14. The non-transitory computer readable medium of claim 8 , further comprising:
identifying a plurality of parameters in outputs generated by the first processing station; and
determining a subset of the plurality of parameters that affect the outputs generated by the first processing station.
15. A system comprising:
a processor;
a deep learning controller; and
a memory having programming instructions stored thereon, which, when executed by the processor, causes the system to perform operations comprising:
receiving a first set of in-process inputs from a first processing station, at the deep learning controller, wherein the first set of in-process inputs are generated at the first processing station deployed in a manufacturing process, wherein the first set of in-process inputs are attributes of the first processing station of a plurality of stations in the manufacturing process;
identifying, by the deep learning controller, a final quality metric of a plurality of final quality metrics for which to optimize the manufacturing process;
generating, by the deep learning controller, an expected value of the final quality metric for an article of manufacture based on the first set of in-process inputs;
determining, by the deep learning controller, based on a comparison of the first set of in-process inputs and the expected value that anomalous activity is present; and
based on the determining, adjusting, by the deep learning controller, downstream parameters of a downstream processing station to cause the final quality metric to address the anomalous activity.
16. The system of claim 15 , wherein the final quality metric is a metric of a completed version of the article of manufacture that cannot be measured until the manufacturing process is complete.
17. The system of claim 15 , wherein the operations further comprise:
adjusting, by the deep learning controller, further downstream parameters of a further downstream processing station.
18. The system of claim 15 , wherein the operations further comprise:
generating, by the deep learning controller, one or more control inputs, wherein the deep learning controller adjusts the downstream parameters of the downstream processing station in accordance with the one or more control inputs.
19. The system of claim 15 , wherein adjusting, by the deep learning controller, the downstream parameters of the downstream processing station comprises:
identifying a key influencer associated with the anomalous activity;
identifying a control input that corresponds to the key influencer; and
adjusting the downstream parameters based on the control input corresponding to the key influencer.
20. The system of claim 15 , wherein the operations further comprise:
identifying a plurality of parameters in outputs generated by the first processing station; and
determining a subset of the plurality of parameters that affect the outputs generated by the first processing station.