Automotive Industry Disruption: A Forecasting Challenge

Rahul Kapoor is a Professor at the Wharton School of Business at the University of Pennsylvania. He is working in collaboration with Professor John Paul MacDuffie to study the disruption in the automotive ecosystem, to which he asks for your participation.

News headlines have recently been emphasizing companies like Tesla and Uber, technologies like self-driving cars, and a greener, more sustainable future through electric vehicles. It’s no secret that the auto industry is in the midst of a dramatic transformation. We’re seeing a changing dynamic from total reliance on internal combustion engines to growing acceptance of electric and hybrid electric vehicles; from driven to driverless cars; from vehicle ownership to vehicle sharing and pay-for-usage ride-hailing — and everything in between. I’ve been fascinated by these changes and am curious how these emerging trends will play out in the next year or so. But I’ve found that it’s even more fascinating to predict the trajectory of these emerging trends by harnessing the wisdom of the crowd. That is, getting people like you – auto enthusiasts who are passionate and eager to learn – to predict industry disruption.

So, my colleague John Paul MacDuffie and I started a forecasting challenge for autonomous and electric vehicles that ran from April to December 2016. We had incredible engagement from the forecasting community – from experienced forecasters to those who are just interested in the mobility space – and we had 1,530 forecasters make over 9,500 predications on 13 questions. The concept of using a forecasting challenge to predict these trends is a new approach in the world of forecasting. Originally, forecasting solely used subject matter experts, and this method was fraught with individual biases and little accountability. Forecasting tournaments (aka challenges) minimize such biases and increases accountability. Now, you may be thinking, what exactly is a forecasting challenge? Forecasting is simply the activity of judging what is likely to happen in the future, based on the information you have now (Cambridge Dictionary). So, in a forecasting challenge like ours, specific questions are asked and answers are predicted by a crowd of forecasters (i.e. you). Answers are judgments that are then translated into probabilities and tracked for accuracy.

Our first forecasting challenge found our forecasters to be highly accurate in their predictions. For electric vehicles, we found that 2016 was far from being a “tipping point” year. There were disappointing sales figures and lack of legislative action in the U.S. However, we saw a positive trend for EVs in that battery costs were dropping relatively quickly. In contrast, with autonomous vehicles, we saw progress – all major players in the space were making significant advances. And policymakers in the US are starting to allow AVs to operate on public roads. More than just accuracy in predictions, forecasters formed a strong community and had extensive interactions on the changing dynamic of the auto industry.

Due to the success of our last challenge, we have recently launched a second challenge to continue studying this exciting space. Like last year’s challenge, we will track developments in technology, automaker strategies, the competitive landscape and the regulatory environment; you will be able to collaborate with the global forecasting community to anticipate the trajectory of an industry in upheaval. We want to encourage you – as an auto enthusiast – to participate in the forecasting challenge. No prior forecasting experience is required and you have the ability to forecast for as many or as few questions as you’d like between now and July 1, 2018. Winners from the challenge will receive a special mention and “badge.” I’m excited to witness the disruption in the auto industry and learn alongside you!

You can join the forecasting challenge here:

Posted in Driverless Car Future, Driverless Car Impacts | Tagged , , , , , | 3 Comments

The Challenge in Establishing Driverless Regulations

It has been a fascinating couple of months monitoring the evolution of the U.S. federal regulations. In early September, the House of Representatives passed an AV bill with unprecedented bi-partisan support (summary found here). Today, the Senate is developing a similar bill that will soon be going to a vote (summary found here). Despite the strong political support occurring at the federal level, there has been significant push-back occurring from the trucking industry (which was ignored in these bills) and the local and state governments (which have been pre-empted from certain responsibilities).

Instead of focusing on the objections, I do think it’s worth celebrating the historical significance of establishing any federal AV regulations at all. Governments around the world are struggling with how to regulate autonomous vehicles, a constantly-evolving technology. The struggle has been: how can the government establish appropriate protections, while still enabling technology advancement/innovation? I don’t think there’s a perfect answer, but I do commend the U.S. for trying! Moreover, this is a positive step towards avoiding the “patchwork of regulation” that’s been occurring at the state level.

Other governments have taken leadership positions as well. The United Kingdom (UK) established a 14-page Code of Practice for testing on UK roads and public places; the code of practice is non-statutory and developed to promote responsible testing. Estonia is allowing driverless vehicles as long as a driver is available to take over control at any time (source). Germany has established ethical guidelines for driverless vehicles, and Japan is also allowing the testing of driverless vehicles with significant stipulations (source). Finally, Singapore has established regulations that include demonstrating “basic roadworthiness and capabilities by passing safety assessments before they are even trialed on roads.” Singapore is also requiring robust accident mitigation plans, licensed drivers, limited test sites, and logging travel data (source).

Clearly, we’re at the earliest stage of regulatory development for driverless vehicles, but let’s celebrate these government agencies for their leadership. Just about every government entity who released regulations acknowledged that they will need to be adapted as the technology is further developed and understood. Let’s also make sure that the government agencies are getting feedback from all relevant stakeholders (private industry, universities, local and state governments, the disabled community, etc).

Are there other regulatory frameworks worth noting?  Any other industries that provide good examples for the driverless vehicles regulations?  Other countries’ regulations worth celebrating?

Posted in Driverless Car Future, Government Considerations | Tagged , , , , , | Leave a comment

Dialogue is the Path to Success

I’m happy to report that Steve Kuciemba is back with another guest blog post. Steve Kuciemba is the National Intelligent Transportation Systems (ITS) Practice Leader for WSP USA.  

The other day I was asked by a public agency client about preparing for the connected & automated vehicle (CAV) revolution.  The question was simple: if you could give 3 pieces of advice what would they be?  Here’s what I told them:

#1 – Expect a Long Transition – we’re not going to wake up one morning and all vehicles will be self-driving, there is going to be a considerable transition period where vehicles operated by computers will coexist alongside vehicles driven by humans.  It could be a period of many years, the mix of vehicle automation could vary by time, location, and technology, and the policy issues will run deep and require time to mature.  This will present a number of safety and mobility challenges, and in the coming years the transportation and technology industries will wrestle with them.

#2 – Be Patient/Flexible – a recent report by the Governor’s Highway Safety Association provided some valuable guidance including a 5-point recommendation to (a) be informed, (b) be a player in your state, (c) understand the role of states, (d) don’t rush into passing laws or establishing regulations, and (e) be flexible – this is a new game.  Those five simple points summarize the state of this fast-moving industry and are great advice for anyone looking to understand the role of state and local governments in CAV.

#3 – Be Ready to Dialogue – when companies begin to discuss testing CAV technology and vehicles in your state or city, everyone should be ready to talk.  You can come up with guidelines, you can make check-lists, you can discuss requirements – but at the end of the day there will be a LOT of unique situations requiring simple communication and conversation.   

This last point is critical.  I’ve talked with a number of State DOT’s and Motor Vehicle Administrations about CAV testing on their roadways, and it’s very tempting to immediately dive into drafting rigorous guidelines and legislative requirements.  Let’s welcome this revolution by first encouraging conversation.  Talk through the issues, discuss the safety implications, and agree to communicate continually as the testing progresses.  

Dialogue is the path to success!

Posted in Driverless Car Development, Government Considerations | Tagged , , , , , | 1 Comment

The Importance of Driverless Pilots

Now that I’ve been at EasyMile for over a month, I’ve recognized how important it is for cities and public transit agencies to develop driverless pilots. The time is now!  The technology is here, but so many other important milestones are still catching up.  Here are just a few of the reasons these pilots are so important:

  • Public acceptance: Call your aunt or uncle and ask if they would be willing to get into a driverless vehicle. Assuming they don’t live in Silicon Valley, a typical response is likely: “Those don’t actually exist” or “No thank you!” This technology has massive safety and mobility benefits and yet public adoption may be one of its biggest deterrents. Pilots present an opportunity for people to see and touch the vehicles….and really process how they could integrate it into their lives.
  • Educate stakeholders: I talk to clients everyday who are sold on the technology and its potential, but their Executive Team, Board, or their operations team still need to be convinced. Pilots present a short-term, affordable way to introduce the technology to decision-makers without any significant commitments (especially since no infrastructure changes are needed).
  • Establish partnerships: The driverless technology is introducing many new players into the mobility ecosystem, including insurance providers, universities, parts suppliers, and privacy/cyber-security experts. The sooner the technology is introduced, the sooner an agency can establish relationships with these new players.
  • Navigate regulations: The regulations are being developed right now! This is the perfect time to introduce the technology and actually influence the regulations at both the federal and state levels. These regulating bodies are receiving a lot of input from the technology developers, but they need to hear from the public agencies as well.
  • Integrate w/ transit: Most would agree that the long-term plan is to incorporate driverless vehicles into transit operations, but most transit agencies are struggling to figure out what that looks like and how that happens. By introducing the technology today, transit agencies can start to see how driverless vehicles can be integrated with their existing operations, fare structure, staffing, etc. And this will also help as agencies struggle to incorporate the driverless technology into their short- and long-range plans.

I can’t say enough good things about driverless pilots!  So cities and transit agencies – what are you waiting for?

Posted in Driverless Car Development, Driverless Car Impacts, Government Considerations | Tagged , , , , , , , , , , | Leave a comment

Possibly the Most Important, Yet Most Forgotten Driverless Mode: the Driverless Shuttle!

The vast majority of the media regarding driverless vehicles is focused on passenger cars. The challenge with this mode is that people expect their personal vehicles to take them everywhere, which requires a Level 5, fully automated driverless vehicle. Right now – the automakers would be thrilled if they could even achieve Level 4 based on this article: “GM has not been as detailed in its plans for the future as Ford and Tesla, who have both said they want to achieve Level 4 autonomy in the next few years.” The reality is that Level 5 is not going to be here anytime soon.

In the meantime, driverless shuttles are actually operating at Level 4 automation TODAY! This means that driverless shuttles can operate autonomously in constrained environments (e.g, campuses, airports, and employment centers). They are a great way to introduce the general public and government stakeholders to the technology while still serving a purpose and they can introduce the driverless technology via shared rides!

Driverless shuttles also present an immediate opportunity for transit agencies to test out the technology and start to incorporate it into their operations. Driverless shuttles can provide first/last mile solutions or circulator services (amongst other examples) and they can be accessible and “green.” As you can see on EasyMile’s website, there are already great examples happening around the world, including in Paris connecting rail stations, in Singapore at a Botanical Garden, and in Arlington, Texas connecting two sport arenas in Arlington, Texas (the latter starting later this year!).

OK, it’s time to admit that I just switched jobs and I now work for EasyMile, a driverless technology company that specializes in driverless shuttles (check out the EZ10 in action). For that reason, this is an entirely biased (yet true!) blog post!  And now you know who to contact if you’re interested in bringing an EZ10 to your venue!

Posted in Driverless Car Development, Government Considerations | Tagged , , , , | 4 Comments

The Unintended Consequences of Driverless Vehicles

We often hear about the really positive consequences that are likely to come from driverless vehicles, including safety and improved mobility, and the negative consequences, including potential increased congestion and loss of jobs, but there are a host of other consequences to consider. In fact, they will likely be significant, even if they aren’t obvious or direct.

Motion Sickness – As this article states, there’s a significant likelihood that people will take their eyes off of the road (since they no longer need to focus on driving) and they will experience motion sickness. Could this create a larger market for motion sickness medicines and therapy?

Decreased Organ Donors – As this article states, if driverless vehicles provide the safety benefits that have been promised, there could be a significant reduction in accidents, which will mean a significant reduction in organ donors. Maybe we can also hope for a significant reduction in the people that need the organs? Let’s hope the medical research community keeps up!

Increased Carpooling and New Communities – As this article states, driverless vehicles will likely significantly increase the potential for car sharing and ride sharing. In addition to (potentially) reducing the congestion on roadways, this concept could create entirely new communities and market potential. Imagine happy hours, AAA meetings, and dates happening within vehicles while also taking people to their destinations.

Municipal Budgets – As this article states, many local governments rely on vehicle-related activities for a large portion of their funding. Sales taxes on vehicle purchases, vehicle registration fees, parking fees, traffic violations, and the list goes on… local governments may want to re-think (and diversify!) their revenue sources since their costs are unlikely to change as drastically.

There are countless other examples! Land use (and parking, in particular), identification cards (no more drivers’ licenses!), shopping/retail completely transformed, driving vehicles as a source of entertainment (think race tracks with driving cars available for rent), and the list goes on!  Please comment and share any other indirect consequences worth considering!

Posted in Driverless Car Impacts | Tagged , , , , , , , , | 6 Comments

Do We Really Need to Own Driverless Cars?

I don’t even want to own a manual car today, so I certainly don’t want to own a driverless car in the future, but will others want to? The car companies would, seemingly, want us to keep buying vehicles (driverless or not), but we know that many of them are diversifying their offerings so that they can sell mobility services (example article). I think it’s worth exploring why people choose to own cars today:

  1. Convenience – With people’s personal car typically parked within 100 feet of where they are at any time, it really is extremely convenient to “hop in” and go directly to their destination, which likely has convenient (and free) parking as well.
  2. Driving Experience – As BMW’s tagline (“the ultimate driving experience”) suggests, people like to drive: steering around curves, driving fast, cruising down highways, etc.
  3. Status – People enjoy the status associated of owning a vehicle and, especially, with owning a high-end name brand vehicle.

I really can’t put cost on that list because owning a car really is not a cost-effective investment. Besides the cost of purchasing a car, people also need to pay for insurance, fuel, license fees, registration fees, taxes, and maintenance. See this article for an estimate of the cost of owning a car.

In a driverless society, people will likely have two choices: own a driverless vehicle or use a range of mobility options. Owning a driverless vehicle will likely cost a few thousand dollars more than the cost of cars today (source), so we know that won’t be a huge deterrent. The next question is: how will our future driverless society fare against our driving values cited above?

  1. Convenience – People may actually find that mobility options (including a mix of driverless fleets of different types of vehicles, public transit, bike share, etc.) to be more convenient – especially if there are enough options, if there is congestion (likely!) and the price is right (less than the cost of owning!).
  2. Driving experience – Well this is going to go away (in theory) no matter what! Dedicated driving aficionados can head to a race course dedicated to driving manual cars!
  3. Status – I believe that companies offering mobility services will find ways to differentiate their offerings in such a way that people will maintain their perceived status (see my last blog post on driverless service differentiators).

Despite all of these points, I do believe there will be people that will continue to want/need to own their own driverless vehicle. So be it. They may even decide to lease their vehicle out when they don’t need it to make some extra cash (check out all the ways you can already lease out your car today when it’s not in use).  What would it take for you to give up owning a driverless car?

Posted in Driverless Car Future | Tagged , , , , , , , | 3 Comments