Traffic noise is an environmental impact of urban highways whose properties depend significantly on statistical properties of the traffic on the highway. Psychological aversion to noise is related to aspects of its variability such as the number, height, and duration of peaks, to the standard deviation and decile range, as well as mean noise level. Some sources of variability on multilane urban freeways include: (a) different emissions within vehicle class among individual vehicles. (b) different emissions among different vehicle classes. (c) formation of clusters or platoons of vehicles of the same type. (d) variation in vehicle spacings within and between platoons. (e) variations in flow among adjacent lanes. Renewal and markov renewal processes provide tractable models for traffic flow. Noise intensity is represented as the noise power contributions summed over all vehicles on an infinitely long, straight, ideal highway; that is, as a linear filter of the multitype point process for vehicle spacings or headways. Moments can be calculated recursively. Some exact theoretical results are available for constant spacings and for exponentially distributed spacings. The exponentially modified skew stable law with exponent 1/2 (i.e. the wiener first passage time density) is a good approximation to the actual distribution. The variability in actual measurements is much greater than in simulations, and much greater than that predicted by the usual design manuals. One simple and effective method for reducing traffic noise variability (hence impact) is a mandatory minimum distance between heavy vehicles in densely populated urban areas. (a) for the covering abstract of the symposium please see irrd abstract no. 224453.
Abstract