S2-2.3 - Traffic Modelling and Control

This paper presents a new agent-based model able to simulate innovative flexible demand responsive transport services, specifically thought to solve the last-mile problem of mass rapid transit. This is particularly needed in areas characterized by insufficient transit supply and lower sprawled demand, where new technologies have the potential to dynamically couple demand with supply. The model compares the performances of two feeder services, one with flexible routes and stops activated by the requests of users, and the other with fixed routes and stops, satisfying the same demand. The case study city is Catania (Italy), where such services could increase the ridership and coverage of a 9 km long metro line that connects the city centre to peripheral areas. Different scenarios have been analysed by comparing a set of key performance indicators based on service coverage and ridership. The first results highlight the validity of the model to identify optimal operation ranges of flexible on-demand services and pave the way for further investigation needed to understand their acceptability and economic viability.

The operation of charging infrastructure is often not profitable yet. However, the integration of charging infrastructure in microgrids enables the introduction of innovative business models, e. g. by using PV generation and storage units. This study presents several business models, i. a. based on self-supply of electricity and smart charging algorithms. These are evaluated by means of real-world transaction data of the EUREF campus microgrid in Berlin, Germany. The dataset comprises data of 3346 transactions of foremost commercial fleet vehicles from the year 2018. The microgrid configurations evaluated in this study involve the integration of PV generation, battery storage, individually as well as combined. A reference is provided by a comparison to the conventional operation of charging infrastructure. The evaluation shows that the integration of PV generation in combination with smart charging is profitable, due to lower expenditures by local energy supply utilization. Additional integration of a stationary battery is less profitable. However, it increases autarky and flexibility by charge during evening and night hours.

Last mile delivery is seen as the most expensive part of the supply chain, and causes significant external effects (e.g. pollution, land use, noise). Therefore, much research has been devoted to finding alternative vehicles to deliver parcels such as drones and cargo bicycles and to finding alternative delivery locations such as parcel lockers. Parcel lockers are seen in most simulation studies as the most sustainable option. However, most of these studies ignore that customers might drive a car to pick up parcels. Also, parcel lockers require space 24/7 whereas a delivery van only requires a parking space for a few minutes while the parcel is handed over. However, during home delivery more space is required for the delivery van while driving due to the increased vehicle kilometer traveled compared with delivery to parcel lockers. The contribution of the study reported in this paper is a simulation tool to evaluate different parcel delivery strategies and customer movements based on real parcel delivery trip data and statistics about the parcel receiving habits. Secondly, the study uses the resulting data to compare the Time-Area requirements of home delivery and parcel lockers.