pilot study: for afcee’s arctic surplus salvage yard · · 2017-05-30for afcee’s arctic...
TRANSCRIPT
PILOT STUDY:
AUTOMATED DATA REVIEW OF SEDD FILES
FOR AFCEE’S ARCTIC SURPLUS SALVAGE YARDREMOVAL ACTION FAIRBANKS, ALASKA
Scott Denzer, Laboratory Data ConsultantsLisa McKenzie, Earth TechAnand Mudambi, EPAJavier Santillan AFCEEJoseph F. Solsky, USACE
AFCEE Chemistry ConferenceAugust 17, 2004
• Over view of SEDD
• Overview of ADR
• The Pilot Study�Design
� Results
• Evaluation of ADR and SEDD
AFCEE Chemistry ConferenceAugust 17, 2004
INTRODUCTION
AFCEE Chemistry ConferenceAugust 17, 2004
SEDD File(Stage 2a, 2b, 3, or 4)
• Stage 2a
• Method Quality Control Data
• Stage 2b
• Instrument Quality Control Data
• Stage 3• Allows for independent calculation of results
• Stage 4
• Captures raw instrument data files
SEDD = Staged Electronic DataDeliverable
AFCEE Chemistry ConferenceAugust 17, 2004
ADR = Automated Data Review
• Developed by Laboratory Data Consultants(LDC) for the USACE
• Laboratory Tool - Allows laboratories toverify EDD completeness and contractcompliance.
• Data Review Tool - Allows data users toperform an automated data review againstproject-specific library measurementquality objectives.
AFCEE Chemistry ConferenceAugust 17, 2004
ADR = Automated Data Review
ADR performs automated review of analyticaldata relative to project requirements.
Project Specs(QAPP/MQOs)
• Methods;
• Target Analytes;
• Reporting Limits;
• Control Limits; and
• Data review criteria.
ADRProjectLibrary
AFCEE Chemistry ConferenceAugust 17, 2004
ADR = Automated Data Review
ADR electronic data deliverable (EDD)comprised of three sets of information reportedin three normalized files
• Result information (A1)• Calibration information (A2 - optional)• Sample information (A3)
AFCEE Chemistry ConferenceAugust 17, 2004
SEDD AND ADR
XMLParser
XMLParser
ADRA1 File
ADRA3 File
SEDD Stage 2b
orStage 3
ADRA1 File
ADRA2 File
ADRA3 File
SEDD Stage 2a • Sample Information
• Result Information
• Sample Information
• Result Information
• Calibration Information
AFCEE Chemistry ConferenceAugust 17, 2004
SEDD/ADR PILOT TEST
Soil Removal Action at Arctic Surplus Salvage Yard ASSY• Phase 1
• Lead and PCBs in soil contaminants of concern
• Data review and validation completed
• Compare results of manual data review to automated data review
• Phase 2
• Semiannual groundwater monitoring
• Pesticides, PCBs, VOCs, and metals contaminants of concern
• Use SEDD/ADR as a tool for data review
• Compare 10% of ADR to manual reviews
AFCEE Chemistry ConferenceAugust 17, 2004
SEDD/ADR PILOT TEST
ADR
ContractComplianceScreening
(CCS)
DataReview
ASSY ProjectSpecifications
from ASSYQAPP
LaboratoryData
SEDD to ADRVVL
Conversion
ASSYProject LibraryDeveloped with
ADR
SEDD Stage 2a
XMLParser
ADRA1 File
ADRA3 File
AFCEE Chemistry ConferenceAugust 17, 2004
PILOT TEST RESULTS – SEDDPARSING
• Logic was required in Parser Tool to limit parsing toa single pass-through.
• DTD validation was required in the Parser Tool toensure the file structure was correct beforeattempting to parse.
• XML technology was/is still evolving. Parser Toolwas designed to take advantage of changingtechnology.
AFCEE Chemistry ConferenceAugust 17, 2004
PILOT TEST RESULTS – SEDDPARSING
�Translator was required in ADR to convertSEDD data content to ADR valid values.
�Project specific requirements must be relayedto the lab before SEDD files are created. ADRproject-specific library is used to relay projectrequirements.
�Working on XML output of reviewed files andADR Library.
AFCEE Chemistry ConferenceAugust 17, 2004
PILOT TEST RESULTS – ADR
Elements Captured in the ADR Review
QAPP Compliance• Reporting Limits• Reporting Units• Target Analytes• Hold Times• Sample Receipt Temperature
Laboratory Batch QC• Assignment of Batches• Surrogates• Method Blanks• Laboratory Control Samples• Duplicates
Field QC Samples• Matrix Spike/Matrix Spike Duplicates• Equipment Blanks• Field Duplicates
AFCEE Chemistry ConferenceAugust 17, 2004
PILOT TEST RESULTS – ADR
Elements Not Captured in the ADR Review
• Calibrations
• Instrument Checks
• Raw Data
• Verification against COCs
• Recalculation of Results
• Completeness of Hard Copy
AFCEE Chemistry ConferenceAugust 17, 2004
PILOT TEST RESULTS – ADR
ADR results compared well to manual reviewresults
– ADR did not miss observations made in manual review– ADR resulted in more flags than manual review– ADR allows for adjustment of flags based on professional
judgment
AFCEE Chemistry ConferenceAugust 17, 2004
PILOT TEST RESULTS – ADR
The automated review flagged the PCB resultswhen the sample receiving temperature was 1.1and 6.2 degrees Celsius. Based on professionaljudgment, the manual reviewer did not flag theresults.
AFCEE Chemistry ConferenceAugust 17, 2004
PILOT TEST RESULTS – ADR
The automated review flagged the PCB results becauseof a matrix spike failure. The manual reviewer did notflag the results because the sample concentration wasgreater than 4x the spiking concentration.
AFCEE Chemistry ConferenceAugust 17, 2004
SEDD/ADR PILOT TESTASSESSMENT
Review and validation time reduced byapproximately 30% assuming:
•10% of the data requires a level III or IV typevalidation
•A complete ADR project library is established bythe project chemist and distributed to thelaboratory prior to sample analyses
•A project chemist oversees the review andvalidation.
AFCEE Chemistry ConferenceAugust 17, 2004
Greater time reductions may be realized:
• If analytical laboratories can capturecalibration data in the SEDD file
• Where level III or IV type validation is notrequired.
• On long-term monitoring and remediationprojects with reduced data validationrequirements and well established qualitycontrol requirements.
SEDD/ADR PILOT TESTASSESSMENT
AFCEE Chemistry ConferenceAugust 17, 2004
SEDD/ADR PILOT TESTCHALLENGES
•Valid values between SEDD and ADR
•LIMS Systems
•Field Sizes
AFCEE Chemistry ConferenceAugust 17, 2004
• SEDD data is easily parsed into existing data reviewsoftware.
• SEDD DTD and ADR CCS ensure quality laboratoryEDDs.
• ADR is a tool for the project chemist to perform 100%review of sample data according to pre-establishedproject library.
• Reduces time and effort to review project data andproduce quality assessment and completenessreports by approximately 30%.
CONCLUSIONS
AFCEE Chemistry ConferenceAugust 17, 2004
• Laboratory Data Consultants, Inc.Richard Amano: [email protected], 760-634-0437
Scott Denzer: [email protected], 760-634-0437
• Earth TechLisa McKenzie: [email protected]
• USEPA
Anand Mudambi: [email protected] 703-603-8796
Website: www.epa.gov/superfund/programs/clp/sedd.htm
CONTACTS