Quality Assurance Protocol: WRAP RMC Emissions Modeling with SMOKE - PowerPoint PPT Presentation

Loading...

PPT – Quality Assurance Protocol: WRAP RMC Emissions Modeling with SMOKE PowerPoint presentation | free to download - id: 25cd24-MmZhN



Loading


The Adobe Flash plugin is needed to view this content

Get the plugin now

View by Category
About This Presentation
Title:

Quality Assurance Protocol: WRAP RMC Emissions Modeling with SMOKE

Description:

Performed by the production modeler, confirmed by the QA manager and PI ... Each step requires modelers initials who checked the step ... – PowerPoint PPT presentation

Number of Views:52
Avg rating:3.0/5.0
Slides: 22
Provided by: UNC5310
Learn more at: http://pah.cert.ucr.edu
Category:

less

Write a Comment
User Comments (0)
Transcript and Presenter's Notes

Title: Quality Assurance Protocol: WRAP RMC Emissions Modeling with SMOKE


1
Quality Assurance Protocol WRAP RMC Emissions
Modeling with SMOKE
  • Zac Adelman
  • Carolina Environmental Program
  • UNC Chapel Hill

Presented October 8, 2003 at UC-Riverside/WRAP RMC
2
Outline
  • Introduction
  • QA Tenets
  • SMOKE Quality Assurance Framework
  • Implementing the framework
  • Specific QA Procedures
  • QA Products

3
Introduction
  • QA protocol provides an explicit framework for
    SMOKE modeling procedures
  • Defines expectations for data quality
  • Allows pre-planning for additional effort
  • Multiple levels of QA
  • Outlines modeling, documentation, and archival
    requirements for thorough QA
  • Facilitates reproducibility of SMOKE results

4
QA Tenets
  • Organization
  • Planning
  • Control
  • Clean
  • Question
  • Document
  • Vigilance

5
SMOKE QA Framework
  • Implementation Three-modeler approach
  • QA Classifications
  • Required Information
  • QA Procedures
  • Templates and Worksheets
  • QA Products

6
SMOKE QA Framework
  • Three-modeler approach
  • Production modeler
  • Data collection, running SMOKE, maintaining
    scripts, creating QA reports, archival,
    delivering SMOKE output, QA documentation
  • QA manager
  • Enforcing adherence to QA protocol, leads QA
    documentation effort
  • Principal investigator
  • Primary technical advisor, maintains project
    schedule, final report writing

7
Three-Modeler QA Approach
8
QA Classifications
  • Model QA Accuracy assurance and problem
    identification
  • System QA Software and data tracking
  • Documentation

9
Required Information
  • Program logs
  • SMOKE QA reports
  • Run scripts
  • Assigns files
  • Inventory summaries
  • QA documents and worksheets

10
Model QA
  • Procedures for determining if SMOKE is working
    correctly and producing expected results
  • Standardized approach for scrutinizing emissions
    model I/O establishes reference points for data
    quality

11
Model QA Components
  • Confirm SMOKE configuration
  • Checking inventories and ancillary input files
  • Confirm SMOKE execution and check logs
  • Confirm that the inventories processed correctly
  • Confirm the information in the SMOKE QA reports
  • Examine indicators for model performance

12
System QA
  • QA process for facilitating reproducibility
  • Accounting system for models and data
  • Begins with installation/benchmarking
  • Tracks all updates/changes to software and data
  • Completes with archival and recording

13
System QA Components
  • Archiving SMOKE source and scripts with CVS
  • Installation freeze points
  • Tracking SMOKE data with QA worksheets

14
Documentation
  • Record results, progress, and key features of
    Model and System QA
  • Central to software and data tracking procedures
  • Tangible record of SMOKE modeling process and how
    to reproduce the results
  • Includes a final report summarizing all modeling
    and QA results

15
Documentation Components
  • Recording SMOKE configuration options
  • Recording SMOKE input/output data
  • Documentation of the QA process
  • Final report
  • Automated project management tool

16
QA Protocol Specifics
  • Step-by-step QA procedures
  • SMOKE program-specific QA checks
  • SMOKE Troubleshooting
  • Procedures for identifying, fixing, and
    documenting emissions problems
  • QA Summaries
  • Procedures for reporting QA process, documenting
    troubleshooting, and compiling a final report

17
Step-by-Step QA Procedures
  • Performed by the production modeler, confirmed by
    the QA manager and PI
  • Specify options, SMOKE flags, and features to
    look for in the scripts, logs, and outputs from
    all SMOKE processors
  • Coincide with a QA checklist for tracking the
    progress of these procedures

18
Example QA Steps
  • Inventory Import (Smkinven)
  • Inventory data cross reference the Smkinven log
    with the completed SMOKE inventory file worksheet
    from Appendix C to confirm that the correct
    inventory files are being modeled
  • Annual, daily, or hourly emissions check that
    the SMOKE setting for importing daily/hourly
    files or annual/daily inventory records is
    consistent with the current inventory

19
SMOKE QA Checklist
  • Detailed explanation of the steps in the
    checklist presented in the body of the protocol
  • Each step requires modelers initials who checked
    the step
  • Steps contain comment fields to answer questions
    associated with the step
  • Final checklist requires modeler, QA manager, and
    PI to sign for certification of completion

20
QA Products
  • Completed Emissions QA checklist
  • Completed QA templates and worksheets
  • SMOKEv2.0 settings
  • Inventory file log
  • Ancillary input file log
  • SMOKEv2.0 execution log
  • SMOKE QA reports and graphics
  • Final report

21
Final Thoughts
  • QA protocol development is an evolving process
  • This protocol presents an experiment in
    micromanaging emissions modeling
  • Experience has shown that in the reality of
    emissions modeling there are no unexpected
    problems expect and prepare!!!
About PowerShow.com