Ingrid Pettersson, Lena Hylving
Traditional manufacturing organizations typically mirror digitally what they produce physically, following the "mirroring hypothesis" . In this article, we take a look at Volvo Car's human-machine interface (HMI) development and how the HMI group broke with established mirrorings when taking on the challenges of digitalization—the organizational aspects of the increasing inclusion of digital technology in physical products and processes.
Well-established car manufacturers today find themselves facing challenges related to digitalization. They have to compete with newcomers as well as online transportation business models; keep up with the swiftly changing movements in consumer electronics that also set customer expectations; and deal with heavily institutionalized development processes and structures. Driver interfaces are becoming increasingly digitalized, displaying complex information from in-car technology. With mirrorings based on the cars' physical architecture, organizations often have a hard time adjusting to digitalization .
Here we look back on the HMI development at Volvo Cars and see how it diverged from rigid development processes in challenging the established mirroring. It was necessary to find new ways of dealing with HMI development at a time of extraordinary advances in technology. This diversion is an ongoing process, started and enabled not only by personal engagement but also by the new tools necessary to make HMIs visible and testable. Increased visualization, as well as the arenas established for cross-organizational collaboration, made the end-user experience of the in-car HMI one of the core priorities of the company. The human-centric brand strategies developed during the process have made their mark on both the organization and the end-user experiences of Volvo Cars.
On the international scene, Volvo Car Group is a small player with high ambitions. The mission: to be the world's most progressive and desired premium car brand and to make people's life less complicated. Currently, the user experience enabled by vehicle HMI is a central part of that mission. That has not always been the case. Looking back 15 years, only a few employees were engaged in HMI development full time, set, as they were, in an organization based on hardware mirroring. The HMI employees got involved as much as they could, but it was difficult to make their voices heard in an organization that did not yet fully understand human-centered design's potential and challenges.
Having developed many innovative safety systems (passive and active) throughout the years, beginning with the three-point seatbelt, Volvo's heart has always been in safety. However, as the complexity of systems increased, and component and process digitalization intensified, it was easy to see that HMI development had to change. Recognizing the need to more effectively communicate with, assist, and influence users had two major consequences: First, the development process needed new tools to transform the complexity into a positive user experience. Second, the organization had to be reorganized in a way that could cope with a digitalized era. Through interviews and observations, we have witnessed how HMI development changed to handle the challenges [2,3].
We focus on two changes made in the development process over the past decade. First, new software-simulation tools were rapidly introduced, shifting the nature of development. Second, a new HMI laboratory was implemented that enabled intensified early testing and joint discussions across the organization. This increased visualization and focus on tangibility led to a new outlook on HMI within the company.
In the development of the Volvo V40 driving interface module, simulation models were sent back and forth between the supplier and Volvo until everyone was satisfied.
Infotainment systems, advanced driver-assistance systems, and driver information all require careful crafting to make them useful, usable, and appreciated. Visualization and iteration are vital for getting it right. Writing down the specification of an interface without first seeing and trying out the interactions would be like painting a picture without looking at the paper. To enable well-founded decisions during the development process, a range of new simulation and prototyping tools were introduced. With the tools came opportunities and a drive to experiment and explore during the development process. A different way of communicating also emerged. In the development of the Volvo V40 driving interface module, simulation models were sent back and forth between the supplier and Volvo until everyone was satisfied. In the past, tedious written documentation would have described the interactions rather than modeled them.
For an interface to be road-ready, developing and evaluating interfaces from a desk do not suffice. The need for a new HMI test facility, including a driving simulator, became apparent. The lab (Figure 1) was finalized in 2009, at a time when key HMI decisions were happening at the company. Those included the introduction of touch technology and fully digital driving information displays, and the reshaping of sound and voice interaction technology. Foundational decisions required a continuous feed of data, so the lab engaged in the ongoing process of innovating, testing, measuring, testing again, and discussing in constant iterations. Being able to use the lab as a place for collaboration, experimentation, and decision making across the organization was fundamental. As such, the lab was located as close as possible to key development teams; HMI developers were on one floor, with the Infotainment and Driver Information software teams on another, all just a short stroll from the Design team. It was adjacent to one of the company's main reception areas, which made it convenient for guests, such as test participants. It was decided that the software must be easy to use by anyone on the HMI team; no education should be needed to set up a basic scenario and get going. However, the laboratory was in no way oversimplified; it contained several integrated systems where data from the simulator software, cockpit signals, cameras, eye-tracking systems, and more was collected by data-measurement units. Examples of initial studies included early explorations of new sound ideas for active safety systems, in which both behaviors and subjective user experience were explored. Other examples included usability and experience studies of speech technology, as well as exploration of new concepts for driver information.
The lab was also a place where the company could evolve and intensify research initiatives with academia and suppliers, with several joint research studies conducted at the venue since its establishment. For example, an early joint research study in the lab explored distraction and the experience effects of touchscreens and pads, as well as provided lessons for the continuous development of the lab . The need for proactive advanced engineering and research projects that keep up with the development of technology, as well as foundational user insights, was gaining recognition. Some of these advanced engineering projects were imported directly into existing car projects, allowing the company to leapfrog certain technological steps. For example, the Volvo V40 driver information interface (Figure 2) originated as an advanced engineering project and quickly became a skunkworks when engineers and designers realized the interface had to make a substantial leap to meet customer expectations and to handle the increased information complexity. It was the first multicolor display in the driver information module, allowing advanced graphics and user choice between personalized themes. By combining a high level of personal engagement, fast changes in processes, the use of new simulation tools, and cross-organizational efforts, the project was delivered at record speed.
The laboratory, the tools, and the advanced research projects sped up the development cycle and improved communication both within the company and with suppliers. The visualizations and simulations that were developed with the tools could be shared with stakeholders who previously had very limited, if any, involvement in HMI development. In this way, the awareness and understanding of what HMI was, and how it could be used, grew within the company. High-level management became more involved and interested in HMI development and began to realize the effectiveness of digital interfaces for communicating the brand to customers. At the same time, the focus on HMI increased, and thus so did the focus for advanced research projects within the area. For example, innovative testing of sound and gesture interactions was explored in terms of emotions and experiences rather than solely from a traditional technological and human factors perspective.
With the new tools, iterative design processes and higher user involvement became further embedded in the organization. While these changes allowed for new ways of working and innovating, they also resulted in more engaged developers who enjoyed their work. This was witnessed in the V40 project, where the HMI manager at the time described her experience: "This project is without doubt the one I am the most proud of, both the product itself ... and the fact that we managed to get a new product-development process in place, which was a lot of fun." The developers tested new ideas in fast loops and tried out new technology continuously, allowing them to learn quickly, constantly improve solutions, and question the traditional technology-driven ways of an automotive company.
At the same time as the technological changes were being implemented for visualization and testing, the organization changed. We observed several departures from existing mirroring as the organization worked toward a more user-centered process. First, new skills, knowledge, and expertise were included in the organization to enable the development of user-centered solutions. For example, people with simulations skills from the mobile industry and fresh graduates from universities were hired, and the scope of HMI was expanded with, for example, sound and voice. This also resulted in engaging interaction experts who worked more profoundly with sound and voice interactions.
When the interfaces became more complex and new stakeholders arose, more communication among developers was required. This was accomplished in more than one way. Some groups physically gathered in the same room to talk and work; new groups were also established. Different ways of working became more acceptable, and the groups who were co-working learned from each other continuously. This made for a more open environment where communication was encouraged and fostered, leading to an improved acceptance of one another's differences. The silo-based organization with high walls between groups began to change as the outcome, rather than internal difference, became the focus. With this cross-organizational work, established mirrorings were challenged, but these changes were complicated.
Tensions emerged for several reasons. For one, with an established mindset of accepting changes in only certain phases of the development process, some had difficulties in appreciating and accepting the possibilities and advantages of constant changes that iterations with simulations, prototyping, and the HMI Lab introduced. Also, as an automotive company with its roots in traditional engineering, the HMI team had difficulty getting subjective measures of user experience accepted as a basis for decision-making, as they lack the exactness and formality of engineering and traditional measuring. Furthermore, the increased attention to HMI development inevitably resulted in less attention to other areas, leading to conflicts.
Several different groups were set up during a short period of time to build the momentum of the new user-centered design vision and to mitigate the tensions. A department named Digital User Experience (DUX) was created. It approached car development solely from an end-user perspective, with the ambition of bringing some balance to the technology-minded organization. The DUX group was fully focused on how to improve the experience of the car, and transportation in general, for the end user. Initiatives that looked beyond the organizational borders of Volvo started up, in several cases with companies outside the automotive industry. DUX is an example of how established mirrorings were ignored and at the same time new mirrorings were reflected when Volvo officially articulated its new focus on user-centered design.
Organizations need to understand how to "break the mirror" with the old in order to establish new mirrorings for the solution they are developing.
Another example of cross-organizational work: feature teams. Again, as the complexity of the product increased, the importance of communication and cooperation between and within different organizational groups became essential, mirroring the complexity of the system architecture and its communication. Initiating user research, such as observing and interviewing users in their homes and cars across different markets, became progressively standardized as the starting point of these feature teams, based on observed needs and continued by communication between interaction designers, function owners, and software and hardware engineers to develop a user-centered feature. As seen in other large organizations , it is not uncomplicated to split user experience and HMI expertise across the organization if a holistic focus on the end-user experience is to remain, and this challenge continues to be debated.
Organizations need to understand how to "break the mirror" with the old in order to establish new mirrorings for the solution they are developing. With institutionalized structures established over a long period of time, such as organizational roles, division of labor, communication channels, and more, the process is very difficult. In the Volvo organization there is still work to be done. However, from this case we have learned that it is necessary to take strong measures in terms of establishing new groups, incorporating new tools and expertise, and daring to divert from set processes and silo-based organizing. Although following these measures might not always yield success, and there are still many challenges left and silos to break, we have seen how the organization has moved from having a few people working on mostly physical interfaces to a whole organization valuing the digital end-user experience.
Between 2013 and 2014, Volvo Cars relaunched its brand with three concept cars that signaled the introduction of Volvo's "Designed Around You" strategy, used for communicating the company's core values to customers and to align work within the company. The strategy focused on human-centric values in the development process, aiming at helping people stay safe and enjoy their journeys with confidence and pride. In early 2014, Volvo launched the new XC90 (Figure 3), reimagining the interior design of the premium SUV with a new driver-control interface. The XC90, now being produced in Sweden, was the first car with Volvo's in-house-developed modular vehicle architecture SPA. Since the launch of the SPA platform, the company was voted as having the "most innovative HMI system" at the 2015 Car HMI Concept & Systems Conference. The XC90 won the 2016 North American Truck of the Year award and has set new sales records .
Volvo Cars has begun moving away from the established institutional processes and static mindsets that mirror the standardized automotive manufacturing logics based on physical components, and instead is moving toward human-centered designs where "Designed Around You" is the motto. Although challenging, the enthusiasm and engagement in the change process has been noticeable. For example, cross-organizational work, a deep belief in user-centeredness, but also experimentation with technology, the introduction of new visualization and simulation tools, and "breaking mirrors" in order to create new structures have been key factors in handling complexity and competition. The technology-driven approach has been balanced and complemented by user-centric approaches, crossing organizational boundaries.
We'd like to encourage companies to break, or at least to question, existing mirroring in order to move forward. Although mirroring seems to be a successful approach when organizing for physical product development , it should be constantly questioned in this fast-changing world where digitalized experiences are becoming the focus of both producers and users. A way to do so, and also to enable holistic UX designs instead of fragmented designs, is to keep HMI visible and tangible and open for experimentation at desktops, labs, and in the field. Product visualizations are not only a way to communicate designs between designers and/or users; they also serve as organizational tools. Having a user-centered mindset in parts of the organization is not enough. It needs to be backed up by management decisions in order to gain momentum. Tools for visualizations and testing, as well as areas in which to meet and discuss, proved crucial for the change in attitude toward human-centered development processes in a large organization.
2. Hylving, L. Digitalizations Dynamics: User Interface Innovation in an Automotive Setting. Dissertation in Information Systems, University of Oslo, 2015; https://www.viktoria.se/sites/default/files/pub/www.viktoria.se/upload/publications/digitalization_dynamics_lena_hylving.pdf
3. Pettersson, I., Hylving, L., Rydström, A., and Gkouskos, D. The drive for new driving interfaces: Researching a driver interface from design intent to end-user Experience. Proc. of the 9th Nordic Conference on Human-Computer Interaction. ACM, 2016, 125.
Ingrid Pettersson is a senior UX analyst at Volvo Car Group, as well as a Ph.D. candidate at Chalmers University of Technology. Her research interest is methodologies for understanding use/rs and UX design. firstname.lastname@example.org
Lena Hylving is a researcher at Viktoria Swedish ICT, where she focuses on organizational changes that come with digitalization transformations. One of her research interests is development processes of HCI and UX in incumbent firms. email@example.com
©2017 ACM 1072-5520/17/05 $15.00
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.
The Digital Library is published by the Association for Computing Machinery. Copyright © 2017 ACM, Inc.