Abstract
The introduction of the concept of a fleet of shared autonomous vehicle (SAVs) which function as a centralized taxi service system presents an innovative way in transport modes. A fleet of SAVs can provide tailored on-demand services via a centralized operation to serve travel demand over time. In our study, An agent-based model (ABM) is developed to simulate tailored time-varying service (TVS) provided by a fleet of SAVs in a demand-responsive fashion. The proposed system can switch the service scheme between the door-to-door service (DDS) and station-to-station service (SSS) automatically based on the time of day; In peak hours, the SAV system aims to serve as many trips as possible with predesignated stations as a SSS by providing an on-demand service. In o-peak hours, a DDS in a demand-responsive fashion is provided by a fleet of SAVs
which can benet the travelers with great convenience. Also, DDS and
SSS provided by SAVs in a demand responsive fashion are simulated separately. The potential benets of TVS provided by SAVs are investigated and then compare it with DDS and SSS. The simulation results indicate that the tailored TVS can increase the utilization of SAVs by 2.5% and number of passengers transported per days by 2.9%. Compared with DDS in peak hours, there are reductions of averaging waiting time and energy consumption up to 25.5 % and 3.7% respectively. In o-peak hour, the TVS can be easily employed to eliminate the avearge 9-minute walking time of the SSS. In addition, we nd out that there is a signicant increase of trips by empty SAVs at around 82% for all service schemes. It is an important issue for further investigation.
which can benet the travelers with great convenience. Also, DDS and
SSS provided by SAVs in a demand responsive fashion are simulated separately. The potential benets of TVS provided by SAVs are investigated and then compare it with DDS and SSS. The simulation results indicate that the tailored TVS can increase the utilization of SAVs by 2.5% and number of passengers transported per days by 2.9%. Compared with DDS in peak hours, there are reductions of averaging waiting time and energy consumption up to 25.5 % and 3.7% respectively. In o-peak hour, the TVS can be easily employed to eliminate the avearge 9-minute walking time of the SSS. In addition, we nd out that there is a signicant increase of trips by empty SAVs at around 82% for all service schemes. It is an important issue for further investigation.
| Original language | English |
|---|---|
| Title of host publication | hEART 2018: 7th Symposium of the European Association for Research in Transportation, 5-7 September, Athens, Greece |
| Pages | 1-8 |
| Number of pages | 8 |
| Publication status | Published - 2018 |
| Event | hEART 2018: 7th Symposium of the European Association for Research in Transportation - , Greece, Athens, Greece Duration: 5 Sept 2018 → 7 Sept 2018 Conference number: 7 https://www.events.tum.de/frontend/index.php?folder_id=877 |
Conference
| Conference | hEART 2018 |
|---|---|
| Abbreviated title | hEART 2018 |
| Country/Territory | Greece |
| City | Athens |
| Period | 5/09/18 → 7/09/18 |
| Internet address |
Keywords
- Shared autonomous vehicles
- on-demand service
- quality of service
- agent-based model
