Ph.D. Thesis Defense Announcement

Developing Correction Factors for Reference Fleet.  The Case of Atlanta I/M Program Evaluation

By

Olga Kemenova

Advisor:

Dr. Michael Rodgers (CEE)

Committee Members:

Dr. Randall Guensler (CEE), Dr. Michael Hunter (CEE), Dr. Catherine Ross (CRP), Dr. Wayne Daley (GTRI/ATAS)

Date & Time: Friday, November 1st, at 9:00 am

Location: Sustainable Education Building (SEB), Room 122

Roadside vehicle remote sensing is a common method to evaluate criteria pollutant emissions from vehicle
fleets. A popular application of this method is the evaluation of the effectiveness of emissions reduction
treatments (e.g. vehicle Inspection and Maintenance (I/M) programs) by comparing the emissions from the
treated (experimental) fleet to those of an untreated (reference) fleet. Since no two large vehicle fleets are
identical, the reference fleet is produced synthetically by application of correction factors. Current federal
guidance assumes that closely located experimental and reference fleets behave similarly if normalized by
model year distribution and Vehicle Miles Traveled (VMT). However inconsistency in the results from these
evaluations questions whether these controls are sufficient to estimate emission reductions attributed to
I/M with an acceptable level of uncertainty.
In addition to age and mileage accumulation, emissions from vehicle fleets can be influenced by
differences in a variety of other factors including differences in fuels and usage patterns and
socioeconomic factors that can influence fleet composition and maintenance trends. This study evaluates
the influence of variety of these factors in determining the reference fleet correction factors for the Macon
and Augusta Georgia fleets for evaluation of the Atlanta, GA I/M program over a twelve-year period from
1998 to 2010. A series of "null" experiments were conducted to evaluate factors that might affect
emissions across vehicle fleets of the same model year distribution. These results showed in addition to
model year and VMT, the reference fleet also required normalization by vehicle type and, in certain cases,
fuel composition to produce a reliable reference fleet correction factors.