Control model

• US 6,260,188 B1
• Filed: 12/31/1998
• Issued: 07/10/2001
• Est. Priority Date: 12/31/1998
• Status: Expired due to Term

• Current Assignee: Kimberly-Clark Worldwide Incorporated (Kimberly-Clark Corporation)
• Original Assignee: Kimberly-Clark Worldwide Incorporated (Kimberly-Clark Corporation)
• Inventors: Lindeke, Anne Marie, Ungpiyakul, Tanakon, Bett, Thomas Arthur
• Primary Examiner(s): Powell, Mark
• Assistant Examiner(s): Holmes, Michael B.
• Application Number:
  US09/224,273
• Time in Patent Office: 922 Days
• Field of Search: 327/170, 318/38, 706/13, 717/1, 717/4, 717/11
• US Class Current: 700/1
• CPC Class Codes: A61F 13/15772 Control

Abstract

Methods of controlling processes having operating conditions, one or more inputs, and one or more outputs. The methods comprise defining acceptable steady state conditions, operating the process, and sampling output on the basis of a dual segment output population model, detecting destabilizing process events which result in output not representing acceptable output parameters, defining operating conditions according to which output is rejected, and defining output parameters according to which output is rejected. Upon detecting a destabilizing process event, the invention temporarily implements sampling assuming a second segment of the output population. The process can include automatically rejecting output whenever any one of either the operating conditions or the output parameters is violated, thereby establishing a leading automatic output rejection and a trailing automatic output rejection in relation to the destabilizing event, and thus defining an automatic output rejection period according to which output is automatically rejected. Such embodiments define a leading transition period proximate and prior to the automatic output rejection period, and a trailing transition period proximate and after the automatic output rejection period. The transition periods, overall, represent periods of higher than average risk that output will be defective. The method contemplates automatically sampling output in at least one of the leading transition period and the trailing transition period.

Claims

1. A method of controlling a process having operating conditions, one or more inputs, and one or more outputs, the process defining parameters of the one or more outputs, the method comprising:

(a) defining a first steady state set of first respective ones of the operating conditions representing one or more previously-specified acceptable operating parameters of the process;

(b) defining a second steady state set of output conditions representing acceptable parameters of the one or more outputs;

(c) operating the process, and routinely sampling output on the basis of a single output population model;

(d) detecting destabilizing process events which result in output not representing acceptable output parameters;

(e) defining one or more third sets of operating conditions according to which output is rejected and/or defining one or more fourth sets of output parameters according to which output is rejected;

(f) upon detecting a destabilizing process event in step (d), temporarily implementing sampling assuming a dual population output, including

1) automatically rejecting output whenever any one of the operating conditions or output parameters of step (e) is violated, thereby establishing a leading automatic output rejection and a trailing automatic output rejection, and thus defining an automatic output rejection period according to which output is automatically rejected in response to the destabilizing event, and

2) defining a leading transition period proximate and before the automatic output rejection period, and a trailing transition period proximate and after the automatic output rejection period, the transition periods, overall, representing periods of higher than average risk that output will be defective; and

(g) identifying output in at least one of the leading transition period and the trailing transition period.

2. A method as in claim 1, including automatically sampling output in at least one of the leading transition period and the trailing transition period, and including returning to routinely sampling output based on a single output population model when both the operating parameters and output conditions indicate the existence of acceptable previously-specified steady state conditions, and no new destabilizing event has been detected.

3. A method as in claim 1, including automatically sampling output from both the leading transition period and the trailing transition period.

4. A method as in claim 1, including evaluating transition period output against previously-specified acceptable output parameters and making a reject/accept decision based on the evaluation.

5. A method as in claim 4 including, when transition period output is rejected, further sampling and manually evaluating output which is farther away, chronologically in time, or by number of units of product, respectively before or after, the automatic rejection period.

6. A method as in claim 1, including manually evaluating transition period output against previously-specified acceptable output parameters and making adjustments to the process based on the evaluation.

7. A method as in claim 1, including operating the process at steady state conditions within the specified acceptable parameters at least one of before a leading transition period or after a trailing transition period, and defining a respective beginning or end of the respective transition period away from the automatic output rejection period.

8. A method as in claim 1, including compiling and evaluating data representing numerous ones of the transition periods and thereby developing desirable adjustments to a length of at least one of the transition periods and correspondingly adjusting the respective transition period in which samples are automatically collected.

9. A method as in claim 8 wherein the adjustment of the respective transition period results in an improved balance between actual accept/reject decisions and the actual acceptability of the output based on the acceptable output parameters.

10. A method as in claim 1, the routine sampling being based on steady state operating conditions.

11. A method as in claim 1, the detected destabilizing events comprising events which have little or no direct affect on operating speed of the process.

12. A method as in claim 1, the detected destabilizing events comprising events which have substantial direct affect on operating speed of the process, including shutting the process down.

13. A method as in claim 1, including sampling transition period output, evaluating the transition period samples, and managing operation of the process based on the results of the evaluation.

15. A method as in claim 1, including automatically rejecting output whenever any one of the operating conditions or output parameters of step (f) is violated, thereby establishing a leading automatic output rejection and a trailing automatic output rejection, and thus defining an automatic output rejection period according to which output is automatically rejected in response to the destabilizing event, and defining a leading transition period proximate and before the automatic output rejection period, and a trailing transition period proximate and after the automatic output rejection period, the transition periods, overall, representing periods of higher than average risk that output will be defective.

16. A method as in claim 15, including automatically sampling output from both the leading transition period and the trailing transition period.

17. A method as in claim 15, including evaluating transition period output against previously-specified acceptable output parameters and making a reject/accept decision based on the evaluation.

18. A method as in claim 17 including, when transition period output is rejected, further sampling and manually evaluating output which is farther away, chronologically in time, or by number of units of product, respectively before or after, the automatic rejection period.

19. A method as in claim 15, including manually evaluating transition period output against previously-specified acceptable output parameters and making adjustments to the process based on the evaluation.

20. A method as in claim 15, including operating the process at steady state conditions within the specified acceptable parameters at least one of before a leading transition period or after a trailing transition period, and defining a respective beginning or end of the respective transition period away from the automatic output rejection period.

21. A method as in claim 15, including compiling and evaluating data representing numerous ones of the transition periods and thereby developing desirable adjustments to a length of at least one of the transition periods and correspondingly adjusting the respective transition period in which samples are automatically collected.

22. A method as in claim 21 wherein the adjustment of the respective transition period results in an improved balance between actual accept/reject decisions and the actual acceptability of the output based on the acceptable output parameters.

25. A method as in claim 15, including sampling transition period output, evaluating the transition period samples, and managing operation of the process based on the results of the evaluation.

14. A method of controlling a process having operating conditions, one or more inputs, and one or more outputs, the process defining parameters of the one or more outputs, the method comprising:

(a) defining a first steady state set of first respective ones of the operating conditions representing one or more previously-specified acceptable operating parameters of the process;

(b) defining a second steady state set of output conditions representing acceptable parameters of the one or more outputs;

(c) operating and controlling the process, and routinely sampling output on the basis of (i) a first output population segment produced under substantially steady state conditions and (ii) at least a second output population segment produced pursuant to destabilizing process events;

(d) routinely sampling output on the basis of steady state operation during steady state operation;

(e) detecting destabilizing process events which result in output not representing acceptable output parameters;

(f) defining one or more third sets of operating conditions according to which output is rejected and/or defining one or more fourth sets of output parameters according to which output is rejected;

(g) upon detecting a destabilizing process event in step (e), temporarily implementing output identification, and operations management, on the basis of the population segment typically produced pursuant to the respective type of extant destabilizing process event; and

(h) when both operating parameters and output conditions indicate existence of acceptable steady state condition, and no new destabilizing event has been detected, returning to sampling output on the basis of steady state operation.

23. A method as in claim 14, the detected destabilizing events comprising events which have little or no direct affect on operating speed of the process.

24. A method as in claim 14, the detected destabilizing events comprising events which have substantial direct affect on operating speed of the process, including shutting the process down.

26. A statistical control model for controlling a process having operating conditions, one or more inputs, and one or more outputs, the process defining parameters of the one or more outputs, the statistical control model comprising:

(a) defining a first steady state set of first respective ones of the operating conditions representing one or more acceptable operating parameters of the process, and target operating parameters included therein;

(b) defining a second steady state set of output conditions representing acceptable parameters of the one or more outputs, and target output parameters included therein;

(c) operating the process, and routinely sampling output on the basis of a single output population model;

(d) detecting destabilizing process events which result in output not representing acceptable output parameters;

(e) defining one or more third sets of operating conditions according to which output is rejected and/or defining one or more fourth sets of output parameters according to which output is rejected;

(f) upon detecting a destabilizing process event in step (d), temporarily implementing sampling assuming a dual population output, including

1) automatically rejecting output whenever any one of the operating conditions or output parameters of step (e) is violated, thereby establishing a leading automatic output rejection and a trailing automatic output rejection, and thus defining an automatic output rejection period according to which output is automatically rejected in response to the destabilizing event, and

2) defining a leading transition period proximate and before the automatic output rejection period, and a trailing transition period proximate and after the automatic output rejection period, the transition periods, overall, representing periods of higher than average risk that output will be defective; and

(g) identifying output in at least one of the leading transition period and the trailing transition period.

27. A statistical control model as in claim 26, including automatically sampling output in at least one of the leading transition period and the trailing transition period, and including returning to routinely sampling output based on a single output population model when both the operating parameters and output conditions indicate the existence of acceptable steady state conditions, and no new destabilizing event has been detected.

28. A statistical control model as in claim 26, including automatically sampling output from both the leading transition period and the trailing transition period.

29. A statistical control model as in claim 26, including evaluating transition period output against the acceptable output parameters and making a reject/accept decision based on the evaluation.

30. A statistical control model as in claim 29 including, when transition period output is rejected, further sampling and manually evaluating output which is farther away, chronologically in time or by number of units of product, respectively before or after, the automatic rejection period.

31. A statistical control model as in claim 26, including manually evaluating transition period output against the acceptable output parameters and making adjustments to the process based on the evaluation.

32. A statistical control model as in claim 26, including operating the process at steady state conditions within the acceptable parameters at least one of before a leading transition period or after a trailing transition period, and defining a respective beginning or end of the respective transition period away from the automatic output rejection period.

33. A statistical control model as in claim 26, including compiling and evaluating data representing numerous ones of the transition periods and thereby developing desirable adjustments to a length of at least one of the transition periods and correspondingly adjusting the respective transition period in which samples are automatically collected.

34. A statistical control model as in claim 33 wherein the adjustment of the respective transition period results in an improved balance between actual accept/reject decisions and the actual acceptability of the output based on the acceptable output parameters.

35. A statistical control model as in claim 26, the routine sampling being based on steady state operating conditions.

36. A statistical control model as in claim 26, the detected destabilizing events comprising events which have little or no direct affect on operating speed of the process.

37. A statistical control model as in claim 26, the detected destabilizing events comprising events which have substantial direct affect on operating speed of the process, including shutting the process down.

38. A statistical control model as in claim 26, including sampling transition period output, evaluating the transition period samples, and managing operation of the process based on the results of the evaluation.

39. A multiple-segment product population, comprising:

(a) a not-culled population segment comprising predominantly acceptable members of the population; and

(b) a culled population segment comprising a steady state cull element, an automatic cull segment, and a transition inspection cull segment, the culled population segment representing

(i) defining a first steady state set of operating conditions representing one or more acceptable operating parameters, and target operating parameters included therein;

(ii) defining a second steady state set of output conditions representing acceptable output parameters, and target output parameters included therein;

(iii) routinely sampling output on the basis of a single output population model;

(iv) detecting destabilizing events which result in output not representing acceptable output;

(v) defining one or more third sets of operating conditions according to which output is culled and/or defining one or more fourth sets of output parameters according to which output is culled;

(vi) upon detecting a destabilizing event in step (iv), temporarily implementing sampling according to a dual population output, including

1) automatically rejecting output whenever any one of the operating conditions or output parameters of step (v) are violated, thereby establishing a leading automatic output cull and a trailing automatic output cull, and thus defining an automatic output cull period according to which output is automatically culled in response to the destabilizing event, and

2) defining a leading transition period proximate and prior to the automatic output rejection period, and a trailing transition period proximate and after the automatic output rejection period, the transition periods, overall, representing periods of higher than average risk that output will be defective; and

(vii) identifying output in at least one of the leading transition period and the trailing transition period.

40. A product population as in claim 39, the transition inspection cull segment representing automatically sampling output in at least one of the leading transition period and the trailing transition period, and returning to routinely sampling output based on a single output population model when both the operating parameters and output conditions indicate the existence of acceptable steady state conditions, and no new destabilizing event has been detected.

41. A product population as in claim 39, the transition inspection cull segment representing automatically sampling output from both the leading transition period and the trailing transition period.

42. A product population as in claim 39, the transition inspection cull segment representing evaluating transition period output against the acceptable output parameters and making a reject/accept decision based on the evaluation.

43. A product population as in claim 42, the transition inspection cull segment representing, when transition period output is rejected, further sampling and manually evaluating output which is farther away, chronologically in time or in number of units of product, respectively before or after, the automatic rejection period.

44. A product population as in claim 39, the transition inspection cull segment representing manually evaluating transition period output against the acceptable output parameters and making adjustments to the process based on the evaluation.

45. A product population as in claim 39, the transition inspection cull segment representing operating the process at steady state conditions within the acceptable parameters at least one of before a leading transition period or after a trailing transition period, and defining a respective beginning or end of the respective transition period away from the automatic output rejection period.

46. A product population as in claim 39, the transition inspection cull segment representing compiling and evaluating data representing numerous ones of the transition periods and thereby developing desirable adjustments to a length of at least one of the transition periods and correspondingly adjusting the respective transition period in which samples are automatically collected.

47. A product population as in claim 39, the routine sampling being based on steady state operating conditions.

48. A product population as in claim 39, the detected destabilizing events comprising events which have little or no direct affect on operating speed of the process.

49. A product population as in claim 39, the detected destabilizing events comprising events which have substantial direct affect on operating speed of the process, including shutting the process down.