Limited Memory Influence Diagrams for Attribute Statistical Process Control with Variable Sample Sizes

Barry R. Cobb
Proceedings of The 11th International Conference on Probabilistic Graphical Models, PMLR 186:1-12, 2022.

Abstract

Limited Memory Influence Diagrams (LIMIDs) are implemented for statistical process control (SPC) to monitor the quality of the output from a production process where the number of defective units in a sample is measured at each time period. The observed defectives provide the input to a decision on whether to stop the process and repair a problematic cause of variation. The model also allows the decision maker to increase the size of the next sample in order to better discern whether or not the process actually requires investigation. The model only requires the user to know the size and result of the current sample to make a decision, in contrast to Bayesian methods that require calculations based on all prior samples and a history of actions. Despite the limited information, the model provides competitive quality costs to existing methods for a wide range of production time horizons.

Cite this Paper


BibTeX
@InProceedings{pmlr-v186-cobb22a, title = {Limited Memory Influence Diagrams for Attribute Statistical Process Control with Variable Sample Sizes}, author = {Cobb, Barry R.}, booktitle = {Proceedings of The 11th International Conference on Probabilistic Graphical Models}, pages = {1--12}, year = {2022}, editor = {Salmerón, Antonio and Rumı́, Rafael}, volume = {186}, series = {Proceedings of Machine Learning Research}, month = {05--07 Oct}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v186/cobb22a/cobb22a.pdf}, url = {https://proceedings.mlr.press/v186/cobb22a.html}, abstract = {Limited Memory Influence Diagrams (LIMIDs) are implemented for statistical process control (SPC) to monitor the quality of the output from a production process where the number of defective units in a sample is measured at each time period. The observed defectives provide the input to a decision on whether to stop the process and repair a problematic cause of variation. The model also allows the decision maker to increase the size of the next sample in order to better discern whether or not the process actually requires investigation. The model only requires the user to know the size and result of the current sample to make a decision, in contrast to Bayesian methods that require calculations based on all prior samples and a history of actions. Despite the limited information, the model provides competitive quality costs to existing methods for a wide range of production time horizons.} }
Endnote
%0 Conference Paper %T Limited Memory Influence Diagrams for Attribute Statistical Process Control with Variable Sample Sizes %A Barry R. Cobb %B Proceedings of The 11th International Conference on Probabilistic Graphical Models %C Proceedings of Machine Learning Research %D 2022 %E Antonio Salmerón %E Rafael Rumı́ %F pmlr-v186-cobb22a %I PMLR %P 1--12 %U https://proceedings.mlr.press/v186/cobb22a.html %V 186 %X Limited Memory Influence Diagrams (LIMIDs) are implemented for statistical process control (SPC) to monitor the quality of the output from a production process where the number of defective units in a sample is measured at each time period. The observed defectives provide the input to a decision on whether to stop the process and repair a problematic cause of variation. The model also allows the decision maker to increase the size of the next sample in order to better discern whether or not the process actually requires investigation. The model only requires the user to know the size and result of the current sample to make a decision, in contrast to Bayesian methods that require calculations based on all prior samples and a history of actions. Despite the limited information, the model provides competitive quality costs to existing methods for a wide range of production time horizons.
APA
Cobb, B.R.. (2022). Limited Memory Influence Diagrams for Attribute Statistical Process Control with Variable Sample Sizes. Proceedings of The 11th International Conference on Probabilistic Graphical Models, in Proceedings of Machine Learning Research 186:1-12 Available from https://proceedings.mlr.press/v186/cobb22a.html.

Related Material