br pi m oshd is the fraction of the
• pi (m/oshd) is the fraction of the aforementioned users who Suramin hexasodium salt choose the transport mode m (known as the “traffic mode-choice sub-model”);
Fig. 2. The traffic forecasting model developed: traffic zoning of the study area with a detail of the inner urban area discretization (a), O/D matrix graphical representation by means of “desire lines” (b).
• pi (k/oshdm) is the fraction of the aforementioned users who choose the trip path k (this is commonly known as the “traffic assignment or path-choice sub-model”).
In order to develop the model a preliminary zoning of the study area was performed: i.e. the study area was discretized into 30 traffic zones where each trip has its origin or destination. In parallel, an in-depth study of the road network of the urban area allowed to develop the traf-fic supply model. The generation sub-model was developed by collecting socio-economic and demographic data from the database of the National Institute for Statistics (available at www.istat.it/it/ archivio/104317), whereas the distribution sub-models (that is usually expressed by means of a “gravitational form”), and the “mode-choice” and “path-choice” sub-models (that are described by “random utility models”) were calibrated and subsequently experimentally validated by performing traffic surveys and traffic counts on several road sections selected in the inner urban area and along the perimeter of the same area (on the “cordon” road sections) between summer and autumn 2015.
In Fig. 2, the traffic zoning of the study area and the Origin/Destina-tion (O/D) matrix graphical layout are reported. The traffic forecasting model was then implemented in a traffic microsimulation open source software (Krajzewicz et al., 2012) in order to capture the kinematics of traffic flow affecting the traffic pollutant emissions.
The vehicles crossing the streets were categorized in four main vehi-cle types due to their different emission characteristics: a) motorcycles,
b) passenger car (50% diesel-fueled and 50% gasoline-fueled), c) diesel-fueled commercial Light Duty Vehicles (LDVs), and d) diesel-fueled Heavy Duty Vehicles (HDVs). The percentages of HDVs travelling along the urban street/road links are equal to 8% and were derived by performing traffic counts on a sample basis, whereas the percentage of HDV in the main rural roads were evaluated by traffic data previously collected by relevant Road Agencies and are equal to 5–12% and 22% for national highways and regional roads, respectively. The EFij data for PM10, CO, CO2, NOx, SO2, As, Cd, Ni, B(a)P, and di-oxins emitted by the different vehicle types (expressed as mg km−1 vehicle−1) were obtained from the database carried out by the European Environmental Agency (2013): the contribution of both ex-
haust ad non-exhaust emissions were considered. The emission factors in terms of particle number (EFN, expressed as part. km−1 vehicle−1) were acquired from the scientific literature (Keogh et al., 2009; Nickel et al., 2013; Wang et al., 2010).
2.2. Evaluation of the particle lung cancer risk emission
The risk related to the airborne particles emitted by the different sources of the city was evaluated by modifying the model proposed by Sze-To et al. (2012). Their original model evaluates the additional or extra risk of developing lung cancer risk (ELCR, excess lifetime cancer risk) of the people exposed to airborne particles carrying carcinogenic compounds. Such ELCR is evaluated by multiplying the dose received by the investigated population and the carcinogenic effects (i.e. “toxic-ity”) of such particles. The model includes the contribution of both sub- and super-micron particles (whereas previous models did not con-sider the sub-micron particle contribution then underestimating the a-priori risk estimate). In particular, the contribution of sub- and super-micron particles was evaluated through the particle surface area and particle mass metrics (PM10), respectively; therefore, the ELCR was cal-culated through the equation: