A multivariate statistical approach to evaluate the effect of after treatment device on bus particulate emissions by in-use testing

The aim of this study is to develop an experimental approach and statistical methodology to assess the effectiveness of technological measures to improve vehicle emissions in real world use. The ultimate purpose is to provide a tool to support decisions made by public administrators and transit authorities, and assess the costs/benefits of environment-oriented investments. In this paper we report some results from the application of the proposed approach within a research project carried out by the Istituto Motori of the National Research Council of Italy, funded by the Ministry of the Environment, in cooperation with urban and extra-urban public transit companies and other public agencies. Our research focused on measuring particulate emissions in real use with and without an after-treatment device. Tests were performed in three Italian cities (Naples, Palermo and Lecce) using seven buses (homologation class EURO 0, 1 and 2) from three public local transit companies. An opacimeter and smoke meter were used to evaluate particulate matter emissions (PM). An original integrated statistical approach is applied to analyse instantaneous data detected on the buses during service operation. Multivariate statistical analysis was used to perform correlations between bus kinematic operating conditions (kinematic sequences) and particulate emissions (opacity sequences).