Methodologies long used in IC, software, and system design are increasingly dominant in a changing automotive supply chain.
MAY 3RD, 2018 - BY: Andrew Macleod and Scott Majdecki
The discipline of automotive E/E systems design is being transformed by trends like electrification and autonomous vehicles, which means there is a premium on methodologies like rapid platform (hardware/ software) prototyping, simulation and test, and electrical architecture optimization. Such methodologies hinge on advanced software design tools and the know-how to make the most of them. This is why design services and consulting are set to matter more than ever.
The change roiling the auto industry is apparent to anyone who pays even passing attention to tech and business headlines. A few months back it was Toyota, one of the world’s largest carmakers, announcing plans to offer electrified vehicles across its entire Toyota and Lexus line-up by 2025, with a goal of selling 5.5 million electrified vehicles annually by 2030. (By way of comparison, the total global market for electrified vehicles was just 3.23 million in 2016.)
When it comes to autonomy, the news if anything is coming at a more frantic pace though it can be tough to separate the signal from the noise. Obviously, the biggest story these days is safety given the handful of accidents, including fatalities, and Tesla’s ongoing wrangling with NTSB. However, announcements also continue apace, from big deals to off-the-beaten-path potential moonshots. An example of the former is Volkswagen’s apparent bid to manage the fleet for Didi, China’s biggest ride-hailing service. And of the latter: a Scottish politician, Ivan McKee, wants his government to set-aside one of the country’s 790 islands for testing self-driving vehicles.
Clearly, when it comes to building cars, the importance of metal bending has been eclipsed by that of creating vehicle electrical systems, which will soon account for 50% of vehicle cost, according to global strategy consulting group, Roland Berger. Average semiconductor content per vehicle is soaring too, projected to go up nearly 25% from 2015 to 2024.
Meanwhile, design cycles are speeding up and product development is changing. In fact, some claim the whole endeavor of creating a car is coming to resemble the chaos of consumer electronics rather the more predictable tick-tock calendar of old-line Rust Belt manufacturing. Our colleague Sjon Moore has a nice whitepaper on this point, explaining how the accelerating design process is straining the V-cycle, which has been honed over decades and is not necessarily amenable to rapid change. And new methodologies long used in IC, software and system design are increasingly dominant in the evolving automotive supply chain. Boston Consulting Group is among those who have noted the overall services trend, saying: “By adopting design and simulation tools, automakers will find that they can significantly improve product development processes.”
This tools+services combination helps shift more of the high value engineering work forward in the design cycle, a trend apparent across all emerging methodologies. For example, when it comes to ICs, siloed supply chains — with fairly rigid and predictable handoffs from SoC vendor to Tier 1 supplier building an ECU to an automaker integrating the ECU into a vehicle architecture — are becoming flat out unsustainable. Automakers are straining under the task of managing multiple ECUs in the same domain, overwhelming software complexity and the general lack of flexibility associated with such an attenuated network of suppliers. What’s needed is a more integrated approach, one that allows for optimizing the overall system of systems performance, efficiency and cost, and doing so as early as possible in the design cycle.
Similar examples are apparent when it comes to components like sensors and batteries, where thermal simulation helps to evaluate sensor placement options and ease ‘range anxiety’ by improving battery life. And the premium on front-loading the design cycle is perhaps nowhere more obvious than in the discipline of EE architecture, facing a stunning amount of complexity due to electrification and self-driving vehicles. With the coming onslaught of dozens of sensors, and hundreds of electric motors and control units, all stitched together by thousands of meters of electric cabling possibly spanning three voltage domains — the value of early stage harness design and optimization has never been higher. Our colleague Puneet Sinha ably walks through various technical domains related to building viable autonomous (and increasingly electric) vehicles in this whitepaper.
Services also have long been essential in minimizing deployment risk associated with new tools, maximizing ROI and thus speeding time to value — virtues never more important than today. No CEO, share price or band value is immune from the cost of missing increasingly ambitious targets for new automotive features, which would have sounded like science fiction even a few short years ago. By the way, when we surveyed auto execs several years back about what they wanted most in design tools, “helps innovation” showed up as third on the list, ahead of items like “improves engineering productivity” and “helps reduce cost.” Our colleague Glenn Reynholds wrote up the results here.
Given the daily onslaught of EV/AV news, we’re probably due to survey execs again. In the meantime, for more on the importance of services and consulting, check out our whitepaper “Tools + services accelerate automotive electrical design”.
Scott Majdecki is director of support and services for the Capital product family at Mentor, a Siemens Business.