Adopting bootstrap for the uncertainty estimation of road traffic noise measurement

In the context of noise pollution, this paper aims to demonstrate the importance of uncertainty evaluation in the environmental noise measurement focusing the attention on the variability of the measurand. Managing the result of a comparison between measured value and the maximum levels permitted in law does not involve a straightforward comparison of values, given that such measurement, which can only be an approximation of the value of the measurand, is expressed as an interval. Consequently, it is essential to take into account the uncertainty associated with the measurement. Attention is focused on the variability of the measurand as a source of uncertainty and a procedure for the evaluation of uncertainty for environmental noise measurement is proposed. Drawing on real traffic noise dataset, the contribution of measurand variability on measurement uncertainty is determined by using the bootstrap method. Experimental results exploring the adoption of the proposed method confirm the reliability of the proposal. It is shown to be very promising with regard to the prediction of expected values and uncertainty of environmental noise when a reduced dataset is considered.