Die Arbeit beschreibt ein Modell zur Abbildung des Verkehrsflusses und von Emissionen heterogener Verkehrsströme. Diese setzen sich aus motorisierten Fahrzeugen wie Pkw, Lkw, Bussen oder Motorrädern, nicht motorisierten Fahrzeugen wie Fahrrädern, aber auch in Europa wenig verbreiteten besonderen Fahrzeugen wie Dreirädern, Rikschas etc. zusammen. Das Modell berücksichtigt Längsbewegungen der Fahrzeuge mittels eines Fahrzeugfolgemodells und Querschnittsbewegungen mittels einer Fuzzy-Logik. Die Emissionen werden fahrzeugtypbezogen und abhängig von der Momentangeschwindigkeit ermittelt, in dem die Strasse in Finite-Elemente unterteilt wird. Zur Berechnung bestimmter Emissionsfaktoren wurde die Methodik in Kenia angewandt, wobei die nicht vorhandenen, empirischen Emissionsdaten abgeschätzt wurden. Es wurde nachgewiesen, dass die besonderen Fahrzeuge überproportionale Schadstoffemissionen und schlechtere Verkehrszustände verursachen. Hierdurch bedingt ergaben sich teilweise Verkehrsstärke-Geschwindigkeits-Diagramme, in denen die erwartete Fundamentalbeziehung nicht mehr erkennbar war. English abstract: This study is concerned with developing a traffic flow and emissions model suitable for heterogeneous traffic streams containing motorised and non-motorised vehicles. Such streams contain vehicle types such as private cars, buses and trucks as well as bicycles, motorcycles and other vehicular forms. The model covers the different vehicle types and allows for peculiar behaviours prevalent in such streams. In addition to an existing deterministic car following rule, the model incorporates detailed lateral movement algorithm with gradual lane change manoeuvres, the decisions of which are governed by fuzzy logic rules. For determination of pollutant emissions, the travel lane was divided into finite elements and total emissions aggregated for each element depending on the vehicle speeds and types. A methodology for estimation of emission rates, based on vehicle maintenance history, for countries without reliable inventories was proposed and used to obtain emission coefficients of CO, HC, NOx and Pb emissions in Kenya. The model was validated analytically and with field data and its predictions were found to be in close agreement with the observed data. It was used to draw relationships between heterogeneity on the one hand and traffic performance on the other, and showed that the presence of the non-standard vehicles in a traffic stream results in higher emissions and deteriorated stream performance characterised by reduced capacities, travel speeds and saturation flows. Moreover, some of the streams had highly scattered volume, speed density plots, which hardly corresponded with the known fundamental traffic relationships. The limited speeds and accelerations of these vehicles and their size, ability to engage in peculiar manoeuvres and individual emission coefficients were identified to be the main determinants of the observed stream performance and emissions. Other unconventional behaviour such as random stopping or partial blockages were found to affect heterogeneous streams only marginally. In conclusion, the model has demonstrated that heterogeneous traffic can be effectively analysed by approaches that take their peculiar characteristics into consideration. In addition, it has also provided new insight in the performance and emission patterns of heterogeneous traffic streams which can be used to formulate regulatory directives governing the usage of lanes or provision of auxiliary carriageways for slower vehicles. (A)
Abstract