No abstract

The Borman Expressway is a heavily traveled 16-mi segment of the Interstate 80/94 freeway through Northwestern Indiana. The Lake and Porter counties through which this expressway passes are designated as particulate matter Ͻ2.5 m (PM 2.5 ) and ozone 8-hr standard nonattainment areas. The Purdue University air quality group has been collecting PM 2.5 , carbon monoxide (CO), wind speed, wind direction, pressure, and temperature data since September 1999. In this work, regression and neural network models were developed for forecasting hourly PM 2.5 and CO concentrations. Time series of PM 2.5 and CO concentrations, traffic data, and meteorological parameters were used for developing the neural network and regression models. The models were compared using a number of statistical quality indicators. Both models had reasonable accuracy in predicting hourly PM 2.5 concentration with coefficient of determination ϳ0.80, root mean square error (RMSE) Ͻ4 g/m 3 , and index of agreement (IA) Ͼ0.90. For CO prediction, both models showed moderate forecasting performance with a coefficient of determination ϳ0.55, RMSE Ͻ0.50 ppm, and IA ϳ0.85. These models are computationally less cumbersome and require less number of predictors as compared with the deterministic models. The availability of real time PM 2.5 and CO forecasts will help highway managers to identify air pollution episodic events beforehand and to determine mitigation strategies.

A stochastic model was developed to estimate the average excess emission of carbon monoxide (CO), volatile organic compounds (VOC), oxides of nitrogen (NOx), and particulate matter (PM2.5) and the traffic delay due to incidents. This work models incident characteristics such as incident clearance time, degree of capacity reduction, and the demand-to-capacity ratio as random variables to derive the statistical characteristics of the excess emissions and traffic delays. It was found that estimated excess CO and traffic delay could be modeled as lognormal distributions. Excess VOC and NOx distributions were found to have the characteristics of a three-parameter lognormal distribution. Excess PM2.5 distribution was found to have gamma distribution characteristics. Average incident clearance time for this study was found to be 26 min. The average degree of capacity reduction and demand-to-capacity ratio were assumed to be 63% and 71%, respectively. An incident with these characteristics is estimated to result in 126.9 kg of excess CO, 20.8 kg of excess VOC, 8.8 kg of excess NOx, 0.27 kg of excess PM2.5 emissions, and 630 vehicle hours of traffic delay. This represents a 138% increase in CO emissions, a 500% increase in VOC emissions, a 26% increase in NOx emissions, and a 43% increase in PM2.5 emissions compared with the normal traffic emissions. Sensitivity analysis of incident management strategies revealed that air pollutant emissions and traffic delay could be reduced by as much as 30% by detouring as little as 5% of the incoming traffic.