ESTIMATING FUEL CONSUMPTION AND VEHICLE TAIL-PIPE EMISSIONS AT SIGNALISED INTERSECTIONS

ABSTRACT: Concern over the negative impact that gasoline powered automobiles have on air quality has prompted renewed interest in methods for quantifying vehicle tailpipe emissions. This paper describes the development of several non-linear regression models that can be used to estimate the additional mass of carbon monoxide, hydrocarbon, and nitrogen oxide that would be expected to be produced by vehicles traversing a roadway, if a traffic signal with known timing parameters was to be installed. One class of regression models uses traffic demands, roadway characteristics, and traffic signal timing parameters as explanatory variables. A second class of models uses number of stops and stopped delay as explanatory variables. The relative merits of these classes of models are discussed. A comparison of the proposed models to similar models contained within the Canadian Capacity Guide indicates marked differences in the relative and absolute impact that traffic signals have on the quantity of pollutants produced. 1