Modelling disruptions in mobility a BP perspective
4 themes for today s discussion 1. What have we published and on this what topic, are and we pursuing what are we for our pursuing internal for needs our internal needs? 2. What has this that meant in terms of model development? 3. What have we learnt so far, and what are we doing at the current time? 4. Sharing What insights of some re assumptions Insights re assumptions and model pivot and pivot points points can we offer? 2
Penetration of Electric Cars 4 2018 BP Energy Outlook BP p.l.c. 2018
Impact of Autonomous Technology 5 2018 BP Energy Outlook BP p.l.c. 2018
Liquid fuel use in cars is broadly flat Changes in liquids demand from cars: 2016-2040 Mb/d 45 40 2.5 Switch to EVs 35 30 22.6 18.2 Other gains in fuel efficiency 25 20 15 2.0 10 18.7 18.6 5 0 2016 Growth in demand for travel Tightening in vehicle efficiency standards Shared mobility EVs 2040 6 2018 BP Energy Outlook BP p.l.c. 2018
4 themes for today s discussion 1. What have we published and what are we pursuing for our internal needs 2. What has this meant in terms of model development? 3. What have we learnt so far, and what are we doing at the current time? 4. Sharing of some Insights re assumptions and pivot points 7
For EO18 we have used a combination of approaches 8
Growth of Vehicle Fleet Cars per 000 pop. 800 600 400 200 0 0 20 40 60 80 GDP / Head ($000) USA (LDVs) Europe China Mexico India Africa Mobility 2050 Model New Demand 2050 model Old Pkm/head ('000s) 25 USA 20 15 EU Mexico 10 China 5 India Africa 0 2010 2020 2030 2040 2050 PKm Growth 9
Vehicle Km are the key pivot Fleet driven Approach Car Ownership (Parc) VKms per vehicle Vehicle Kms Load factors Travel by Car Modes (Pkm) Mobility driven Approach Define the Energy Outcomes 10
Our thinking has 3 foundations mega 9 trends 1 11
Our thinking has 3 foundations 2 What drives consumer choice on Mode of Travel, and Vehicles? PKm Mode Choice Shared? VKm Mode Autono mous + Energy Vehicle Choice Electric or ICE 12
Our thinking has 3 foundations Quality With a need to assess Cost vs Convenience 3 Consumer Trade offs Cost Time 13
Our thinking has 3 foundations mega 9 trends Quality PKm VKm Energy Consumer Trade offs Mode Choice Mode + Vehicle Choice Cost Time Shared? Autono mous Electric or ICE Choice modelling on Vehicles, and Modes 14
Mobility 2050 uses two consumer choice models to estimate mode share & quantify the impact & dimensions of new mobility for BP. 1 Mobility Activity Passenger Kms - History - Regression Model - Projections 2 Consumer Choice Car Choice - Owned Car + - Shared Car Mode Choice - Owned Car - Shared Car - Rideshare - Bus - Train - 2/3W 3 PKms by Mode Vehicle Kms Fuel Demand 4 Vehicle Sales & Parc Diffusion curves for adoption of Sharing & AV technology 15
4 themes for today s discussion 1. What have we published and what are we pursuing for our internal needs 2. What has this meant in terms of model development? 3. What have we learnt so far, and what are we doing at the current time? 4. What insights re assumptions and pivot points can we offer? 16
Initial learnings & current programme For now much of the old paradigm remains valid, but we are absolutely convinced about the need to reflect new mobility into our approach to long term modelling for transportation energy. A clear challenge is to manage the evolution within our modelling. Personal mobility expands the modelling scope, requiring more detail & mode switch interactions to be considered. There is a complexity trade-off; the scale of changes could be very significant, particularly over the long term, for energy demand. New mobility requires thinking beyond techno-economic choices. It clearly involves behavioural-economic choices but the evidence for the strength of some changes is still building as such this inevitably requires scenario & sensitivity tools within the modelling framework. We built a new model, that is sizeable & complex, and we are happy with what we achieved with it, but we need to still be open to adapting our approach, and we will actively monitor alternative ones 17
Initial learnings & current programme Phase 2 we are clearly moving on from the initial development, into model iteration & continuous improvement We are currently taking on board the challenge to integrate our two models into a single model We are clearly wanting to test & improve the key assumptions, as well as widen our data sources & capture 18
4 themes for today s discussion 1. What have we published and what are we pursuing for our internal needs 2. What has this meant in terms of model development? 3. What have we learnt so far, and what are we doing at the current time? 4. What insights re assumptions and model pivot points can we offer? 19
Insights on Assumptions & Pivot Points Consumer vs Regulatory-driven world? There is no certainty, we believe that there is a need to consider & model both kinds of outcome for some time. Vehicle ownership is better understood than personal mobility our experience is that the latter requires understanding of more factors, with more detail so it is necessary to be prepared to develop / research the necessary inputs & assumptions, particularly for a global model, that covers the entire range of settings / energy demands. Our initial focus has been on the movement of people as opposed to the movement of goods, but we recognise need to go there too, it is likely to require another development project (The various aspects of) Consumer choice, and technology adoption are key areas We believe the value of time is critical, and shapes the outcomes Moreover, there is a need to think carefully about the diffusion parameters, what AV might entail, and when will the market be ready for it? 20
Some takeaways from our analysis Trillions 100 80 60 40 20 100% 80% 60% 40% 20% World - Passenger kilometres 0 2015 2020 2025 2030 2035 2040 2045 World - Passenger kilometres 0% 2015 2020 2025 2030 2035 2040 2045 Bus Rail 2/3 wheeler Car - Rideshare Car - Taxi / Shared Car - Drive Self Bus Rail 2/3 wheeler Car - Rideshare Car - Taxi / Shared Car - Drive Self Autonomous driving may be a stronger vector for change than powertrain electrification Autonomy supports electrification particularly for Shared Car-based Transportation modes The future regulatory ecosystem (for AVs, in Cities, on highways etc.) is emerging but not certain For now, technology development is leading, and Consumer Adoption is likely to lag uncertainties lie in timing and pace of the take off The convenience of personal, owned cars may be more enduring than some suggest 21