Early Warning Bells for Autonomous Technology
June 29, 2017
By Kristin Kolodge
The iconic J.D. Power Initial Quality StudySM (IQS), the most widely watched measure of automotive quality in the world, was released in June, celebrating three decades of measuring consumer satisfaction of new cars. The quality landscape has certainly changed since the study’s inception in 1987. Overall quality has improved at an accelerating rate over the past three years as automakers have come to grips with some of the larger consumer complaints. This has resulted in new-vehicle quality that is at its highest level ever, improving a significant 8% from last year. Despite higher technology content on vehicles increasing the chance that something could go wrong, fewer problems per 100 vehicles (PP100) are being reported by consumers. Kudos to auto manufacturers on such an accomplishment.
The only IQS category to worsen this year is Features, Controls and Displays (FCD). The largest increase in problems came from driver assistance systems: cruise control (primarily adaptive cruise); lane departure warning; collision avoidance/alert systems; and blind spot warning. These features comprise some of the building blocks of autonomous vehicles, and an increasing number of consumer-reported problems sounds warning bells for automakers and suppliers. Consumers will need to be convinced that these systems are foolproof before they will give up driving control to autonomous vehicles.
Initial Quality Study - Total Features, Controls and Displays (FCD)
Problems per 100 Vehicles (PP100) • Lower Score = Higher Quality
Is such an uptick really something to be concerned about? Based on history, yes. (Think voice recognition and Bluetooth pairing.) But the sensitive path to autonomous vehicles adds another layer of worry.
Today’s driver assistance features are at a level where they are meant to do just that—assist—sometimes through warnings and other times by implementing a level of automatic braking or steering. This is what consumers have shown an overwhelming interest in and are willing to pay for, based on our research.
Now they are getting what they asked for. However, the result is an increase in consumers raising their hands, saying the systems are difficult to use or understand (DTU). These types of systems are unlike anything the driver has experienced before, so their mental model of how it should work has not formed. This makes it all the more critical for a solid human machine interface (HMI) design that ensures the technology is simple, intuitive, easy to use and useful from its first interaction point.
However, these driver assistance systems can conceptually be scary to use the first time because of a lack of understanding about how to use it; what the driver’s role should be; and the level of trust that it will work. There’s also the potentially serious consequence of something going wrong.
J.D. Power has stated the importance of a positive first experience with these lower level automated systems. The experience will determine if the trajectory to accepting higher levels of automation—including full self-driving vehicles—will be expedited or hindered.
What IQS tells us this year is:
- Cruise control DTU issues attributed to adaptive cruise control have doubled, now accounting for 34% of all cruise control DTU problems.
- Self-parking penetration is still low (6%), though it’s up from 4% in 2016. Of note is the increase in DTU problems at 2.41 PP100 vs. 1.66 in 2016 with this technology. Considering the amount of vehicle automated control this system has, it suggests execution issues of those aforementioned HMI principles.
- There is substantial penetration opportunity for blind spot warning, lane departure warning and collision avoidance/alert systems. All of these show year-over-year growth of 10 percentage points or more.
With increasing penetration comes an inherent potential for more quality issues. The source of the quality issue matters, whether it is by design or it is a product defect. These driver assistance systems are evidence that the source of the issue is DTU, which is directly linked to the system design. DTU issues are very challenging to resolve because the user experience is rooted in the software and design architecture of the vehicle. However, it may be an opportunity for manufacturers to flex their growing over-the-air update muscle to improve the software where it is possible more frequently.
It is in the best interest of every manufacturer to get drivers to use these driver assistance systems and build their reliance on them. Doing so will inspire the next evolution, or what the industry will call the next level of automation.
Until then, the warning has been sounded.
# # #
Kristin Kolodge is executive director for driver interaction and HMI at J.D. Power. For more than 20 years, she has been at the center of reducing the number of HMI alarm bells for drivers, and is working to prevent autonomous vehicles from becoming a user-experience debacle similar to that of voice recognition.
The information contained herein has been obtained by J.D. Power from sources believed to be reliable. However, because of the possibility of human or mechanical error by our sources, J.D. Power does not guarantee the accuracy, adequacy, or completeness of any information and is not responsible for any errors or omissions or for the results obtained from use of such information.
This material is the property of J.D. Power or is licensed to J.D. Power. This material may only be reproduced, transmitted, excerpted, distributed or commingled with other information, with the express written permission of J.D. Power. The user of this material shall not edit, modify, or alter any portion. Requests for use may be submitted to firstname.lastname@example.org. Any material quoted from this publication must be attributed to “J.D. Power Mobility Disruptors, © 2017 J.D. Power. All Rights Reserved.” Advertising claims cannot be based on information published in this report.