Autonomous Vehicle Impacts on Traffic and Transport Planning

Transcription

1 Autonomous Vehicle Impacts on Traffic and Transport Planning AITPM Regional Seminar 11 May 2017

2 Discussion Paper

3 AVs/CAVs Autonomous/Automated Vehicles (AVs) Connected and Automated Vehicles (CAVs) Increased safety/reduced accidents Over 90% of vehicle crashes cites human error as the cause Enhancement of travel experience Do other activities instead of focusing on the road Reduce parking requirements Re-allocate road space to other users Improved mobility for all

4 Levels of Automation

5 Transport Planning Current Practice Short, medium and long term horizons Traffic and transport demand modelling based on land use, planned road network, public transport infrastructure, parking, etc In the ACT (Canberra Strategic Transport Model): Land use: Population, Employment, Retail, School and Tertiary Enrolments Road capacity: derived from Austroads and Highway Capacity Manual (HCM) Generalised Cost of Travel: Cars, Public Transport

6 Future Transport Ecosystem AV/CAV Forecasts: Level 4 available by 2025, 40-60% fleet penetration in the 2050s (conservative estimate) Mobility as a Service (MaaS)/Shared Autonomy Electric Vehicles Freight (autonomous trucks and truck platooning) ACT medium/long term plans: 2031/2041 Travel demand forecasting and transport modelling vehicle technology and driver /traveller behaviour will be very different Update transport planning assumptions and model input parameters

7 Road Capacity Minimal to no impact in the interim; possible reduction in the early stages Significant capacity improvements in the longer term Rate of change of improvement goes much faster at % CAV penetration *Vehicle-to-vehicle (V2V) communications critical for improved road efficiency (Tientrakool, et al, 2011)

8 Sensitivity Testing: Arterial Road Capacity 10% Cap AR increase 0.7 km/h increase Improves average speed of cars Rate of improvement slows down as capacity increases At 100% increase, average speed is improved by almost 8 km/h Gradual decrease of VHT as capacity is increased

9 Cost of Travel (Current) Cost of Travel Cars: Fuel cost + VOC + in-vehicle time cost + parking cost (if applicable) PT: PT fare + walk/wait/transfer time costs + in-vehicle time cost Time cost = Value of Time (VOT) In the current CSTM: Future vehicles = today s vehicles Fuel cost and VOC Petrol cars Parking cost: increases over time Public transport is generally the same transit modes as today VOT is assumed not to change in the future

10 Cost of Travel (with AVs/CAVs) Significant proportion of Electric Vehicles (EVs) Fuel cost and VOC calculations need to be revisited Significantly reduced parking requirements Parking costs could be insignificant Potentially new forms of public transport (personal mobility service, shared autonomy) Calculation of PT costs could be considerably different Opportunity to engage in other activities (e.g. work, videos, sleep) while travelling by car VOT in AVs would be lower than VOT in non-autonomous vehicles

11 Sensitivity Testing: Value of Time (VOT) vs Speed 10% VOT decrease 1.4 km/h decrease Increase of car demand on roads, hence the average speed reduction Average speed at 50% VOT + 100% AR capacity > Average speed at current VOT and AR capacity

12 Trip Making Increased convenience and accessibility additional travel, more vehicles on roads New forms of car (or CAV) travel Car trips by people who cannot drive (e.g. children, PWD, the elderly) CAV-induced trips (i.e. trips that would not have been made by people who can drive if CAVs were not available) Zero-occupant car trips Additional demand could offset the capacity benefits; V/C may not change Increase in Vehicle Kilometres Travelled (VKT) and Vehicle Hours Travelled (VHT) people not minding longer travel distances

13 Sensitivity Testing: Value of Time (VOT) vs Trip Length Reducing VOT increases average car trip lengths 50% VOT reduction 353 metres average trip length increase

14 Parking Reduced demand for car parks A single Shared Autonomous Vehicle (SAV) can replace 12 privately owned vehicles and eliminate 11 parking spaces (Fagnant and Kockelman, 2014) Planned car parks can be re-planned and existing car parks can be re-developed Parking charges no longer a concern Parking patterns will be altered Moving a parking space from the CBD to a suburban area could save around $US3,000 (Fagnant and Kockelman, 2014)

15 Suburbs Suburbs Suburbs CBD Suburbs Car Park

16 Public Transport Road capacity increase PT mode share reduction 100% capacity increase 0.8% PT mode share decrease Reducing VOT sharper decline in PT mode share 50% VOT reduction 2.3% PT mode share decrease

17 PT Impact Scenarios Complementary

18 PT Impact Scenarios Competing

19 Concluding Statements Driverless technology is happening sooner than what most people think Significant fleet penetration by 2030s or 2040s Preliminary sensitivity tests indicate that small/simple changes in the model (based on possible impacts of CAVs) could result in considerably different forecasts from what they are now Worth conducting further investigations (e.g. traffic & transport modelling) Governments need to take this technology into consideration when developing long term urban and transport plans Review current practices and investigate ways to incorporate CAVs in the planning process NOW is the best time/opportunity to develop planning tools and processes in preparation for a very different future transport environment

20 Recommendations for Future Discussions Transport planning and modelling processes: review and modify Road upgrade plans: review and possibly re-prioritise Public transport planning: future-proofing of long term plans Parking: revisit current plans and guidelines Integrated Transport Planning: nationally recognised set of guidelines that include AVs/CAVs Further modelling and simulation of autonomous vehicles: to better understand the extent of network benefits (or disbenefits) On-road trials in the ACT?

21 We always overestimate the change that will occur in the next two years and underestimate the change that will occur in the next ten. Don't let yourself be lulled into inaction. -- Bill Gates --