The Fuse

Study Shows the Unexpected Benefits of Autonomous Vehicles

by Paul Ruiz | @pmruiz | February 27, 2015

It’s Monday morning, and you’ve just arrived at work. You step out of your car and tell it to “Go home.” It follows the instruction, then takes your children to school, picks them up at the end of the school day, brings them home, and then returns for you when you’re finished at the office. Sound like science fiction? Not according to a new study by the University of Michigan’s Transportation Research Institute (UMTRI).

Researchers Brandon Schoettle and Michael Sivak conducted an analysis of the Federal Highway Administration’s latest National Household Travel Survey. They found that vehicle automation and, in particular, the potential for cars to use self-driving “return-to-home” capabilities, could significantly reduce overlap in average vehicle use per household, and over time, lead to a decrease in car ownership rates by 43 percent nationwide (from 2.1 cars per home to 1.2).

The UMTRI study reveals several important facts about car ownership in America. First, many families own two or more vehicles, largely because it is impossible for a single car to be used by multiple individuals in a given day. During the daytime hours, most cars sit unused in parking lots and garages. The story is essentially the same at night. A 2012 survey by the International Parking Institute, an industry group, found that cars spend 90 percent of their lives parked. Car sharing initiatives have demonstrated the ability of multiple-occupancy vehicles to ease urban traffic congestion and create household fuel savings. But there has been little scholarly research assessing the potential for vehicle automation on demand for vehicles per household.

Vehicle automation may reduce the number of cars on the road. This would have clear effects on household vehicle use, the researchers say. Using the National Household Travel Survey, an aggregate data set that logs travel times and destinations, Schoettle and Sivak found that 84 percent of passenger car trips do not conflict or overlap. Yet, 42 percent of U.S. households currently own two or more vehicles out of necessity for different travel destinations. A fleet of self-driving cars could lower that number to 15 percent. Less than two percent of households would require three or more cars, from 27 percent today.

A bigger question is whether or not future advancements in vehicle automation will result in overall fuel savings. On its face, fewer vehicles per household would seem to cut consumption but does not translate directly into fewer vehicle miles traveled (VMT). Even in our earlier example, a vehicle that transports a parent to work, then travels home to transport children to school, before traveling home for the day, then traveling back to school to pick up children, back home to drop the children off, and then travelling to the parent’s work in order to bring them home, is in fact traveling twice as far as two vehicles that travel to each destination and back once in the same day (the vehicle takes two trips to school and two trips to work). However, the potential increase in travel demand extends far beyond this phenomenon. Fully automated vehicles could fundamentally change access to transportation systems, giving virtually everyone access to a car, including the very young and very old. VMT for passenger vehicles, which fell in the years following the recession, could tick upward once more, with some suggesting that automation may diminish demand for public transit.

However, despite the likely uptick in demand, fuel savings are still possible. In 2013, research from Morgan Stanley reported that the elimination of human error and inefficiency could save 20 to 30 percent in fuel costs. Related improvements in route efficiency, cruise control, and new engine and transmission technologies, they say, could save an estimated $158 billion. Additionally, intelligent automated systems could avoid congested traffic areas and costly accidents, perhaps saving 2.9 billion gallons of fuel.

When consumer surveys ask Americans to rank the important feature of vehicles, fuel economy is frequently cited as an important consideration. Accordingly, Schoettle and Sivak’s prior research has unveiled tremendous popular support for fuel economy savings that could result from vehicle automation. Seventy percent of respondents in their October 2014 survey—more than any other segment—said autonomous vehicles will improve fuel economy, more than those who said self-driving cars will reduce the probability of getting into accidents, cut traffic congestion, lower travel times, or decrease insurance rates. Yet, only 56 percent say they have favorable opinions of the technology. A Q4 2013 Navigant Research study found similar levels of favorability for electric vehicles—another burgeoning technology—among American consumers. In that survey, hybrid cars earned 67 percent favorability and fully electric cars earned a 61 percent favorable rating.

The reality is that new technological advancements require consumer adaptation. The UMTRI research, which suggests that households could do more with less, relies on the presumption that consumers will be amenable to owning fewer vehicles per household.

Technologically, autonomous vehicles may be better equipped to handle the transition to battery electric cars. In traditional internal combustion engine vehicles, for example, separate and distinct systems, like the vehicles’ powertrain, safety features, and media components, do not communicate with one another. This is not the case in electric vehicles, which operate more cohesively as a single, unified system. A fully integrated autonomous vehicle would require constant communication between its internal and external components, which explains why many early autonomous vehicle prototypes have also been fully electric. Furthermore, the added flexibility of motor and battery placement in an electric vehicle allows for much greater diversity in design options than petroleum-powered counterparts.

The car of the future may be intelligent, integrated, and efficient, but getting there is largely a question of consumer attitudes. Furthermore, the specter of increased travel demand could undermine many of potential benefits for reducing fuel consumption. Therefore, a tight link between alternative fuel technologies and autonomous drivetrains will be essential in order to ensure that this emerging technology doesn’t ultimately detract from efforts to reduce U.S. oil dependence.