Download presentation

Presentation is loading. Please wait.

Published byTamia Brisley Modified about 1 year ago

1
IIIG LTMS V2 Review

2
LTMS V2 Review Data Summary: – Includes 285 Chartable reference oil results from all test laboratories – Most recent chartable reference oil result included in data set is March 22, 2010 – Includes all ACLW data that is currently exhibiting a mild trend – WPD and PVIS defined as “Primary Parameters” – ACLW defined as “Secondary Parameters”

3
LTMS V2 Review Proposed Limits for IIIG LTMS v2 example: – Limits for e i & z i : Shewhart Chart of Prediction Error e i = Y i – Z i-1 Limit TypeLimit* Level Level Level EWMA of Standardized Test Result Z i = λ(Y i ) + (1 – λ)Z i-1 Limit TypeλLimit Level 2 Upper Limit0.2 TBD by SP Input Level 2 Lower Limit0.2 TBD by SP Input Level 10.20

4
IIIG calibration attempt summary: – Of the 289 total, 73.3% acceptable, 17.3% failed acceptance, and 9.4% were invalid

5
IIIG Alarm Summary (all labs & chartable results): – Below summarizes the unacceptable, reduced, and extended reference interval count for LTMS v1 & v2.

6
IIIG Alarm Summary (all labs & chartable results): – Below summarizes the unacceptable, reduced, and extended reference interval percentage for LTMS v1 & v2.

7
IIIG Alarm Summary (by lab): – Below summarizes the unacceptable, reduced, and extended reference interval count for LTMS v1 & v2 by test lab (285 Chartable test results).

8
IIIG RMSE for the PVIS parameter: – RMSE calculation is a function of the average deviation from the target and within lab variation.

9
IIIG Relative Pass limit of a Candidate test for the PVIS parameter:

10
IIIG RMSE for the WPD parameter: – RMSE calculation is a function of the average deviation from the target and within lab variation.

11
IIIG Relative Pass limit of a Candidate test for the WPD parameter (with a [GF-4] 3.5 limit):

12
IIIG RMSE for the ACLW parameter: – RMSE calculation is a function of the average deviation from target and within lab variation.

13
IIIG Relative Pass limit of a Candidate test for the ACLW parameter:

14
Undue Influence example for TPVIS (e i ) data – Plot of e i data with no Undue Influence adjustment

15
Undue Influence example for TPVIS (e i ) data – Circled results with “capped” adjustment (at limit) Result adjusted if |Y i – Z i-1 | > and |Y i – Y i+1 | > 2.066

16
Undue Influence “Capped” Result Summary – (285 Total Chartable results)

17
Appendix LTMS V2 Charts By Test Lab

18
Lab A

19

20

21

22
Lab B

23

24

25

26
Lab D

27

28

29

30
Lab E

31

32

33

34
Lab F

35

36

37

38
Lab G

39

40

41

Similar presentations

© 2016 SlidePlayer.com Inc.

All rights reserved.

Ads by Google