Volvo Car may be a model for automotive manufacturing in the near future. Faced with unique circumstances since the sale of the company by Ford in 2010 to Zhejiang Geely Holding (Geely Holding) of China, their response is instructive. In 2013 the company sold over 425,000 units spread over eight base models. That volume gives them a miniscule 0.5% market share.
“We would like to have 800,000 units per year to make a sustainable company,” said Anders Carlsson, Senior Engineering Advisor for Volvo Cars, acknowledging that profitability, rather than market share, is what matters.
Like any successful company, Volvo needs differentiation. Creating a driver-centric experience has always been important to Volvo, according Carlsson. This initially led the company to emphasize safety. Now, they are looking to an internet-connected experience.
“We will have fully integrated iPhone and Android compatibility capabilities in our car,” he explained. “We will actually connect your phone into the dashboard and you will see it in a big screen in the center console.”
At the same time, there are many elements in Volvo cars that need not stand out. To deliver on features that only need to be competitive—rather than outstanding—is where the rapid-follower business objective enters the picture. To accomplish this requires reducing time to market, from concept to start-of-production. When discussing automotive product development lead times, different numbers are bandied about. This is usually because the speakers are talking about different start times, from initial concept to final engineering release of drawings to start of tooling.
Carlsson made it clear during the 2014 CIMdata PLM Road Map conference in Plymouth, MI, that Volvo's standard of comparison starts at Concept Development. This is when an all-new vehicle resides in a marketing white paper and maybe some initial sketches. “The cars we are launching now, like the new XC90, have been in development for around 42 months,” he explained.
The goal is that by 2020 Volvo will reduce that lead time to 20 months, from concept to start-of-production.
An intermediate goal is to reduce the lead time to 30 months by 2015 — still a remarkably ambitious goal. “This proposal sends a message to the community and to our organization that we need to start to think differently, and this is coming from all the highest authorities in our company,” he said.
Virtual tools replacing physical tests
As is increasingly common in the industry, Volvo is turning to virtual testing and CAE to replace time-consuming physical tests. Carlsson acknowledged that Volvo has been moving in this direction for the last 15 years with varying results. But the competitive and business pressure coupled with today’s accurate, high-fidelity CAE simulations means Volvo will eliminate physical verifications on complete vehicles.
“The first car that they are actually going to build in the plant will be the first car that you actually see drive,” he said. This eliminates an entire category of prototypes common in the industry. Such prototypes have different names depending on the company; Volvo labeled theirs Verification Prototypes.
Carlsson predicts in 2017, the first Volvo cars that have not been tested as whole vehicles will roll off the line. Subsystem and special mule-rig tests will remain, however. “This is a big thing for us,” he said. In the author’s own experience, this can eliminate 9 to 12 months of development time — a big win on the way to a 20-month process.
The key is confidence in CAE predictions of vehicle driving dynamics, crash, and assembly operations. Carlsson presented materials showing that a relatively new CAE exchange format, the Functional Mock-up Interface (FMI), was an important enabler in coupling tools together and exchanging model data for a complete vehicle simulation.
Common architecture, decoupled engineering
However, eliminating the physical Verification Prototypes will only get them to 30 months. “This is the limit of what you can do with the technology that we have,” explained Carlsson. The next step in Volvo's development is to look at the underlying vehicle and engine architectures.
“We have been talking about platforms for quite some time in the car industry, for the last 20 to 30 years, but now we’re talking about architectures and modules,” he noted. The XC90 is the first Volvo developed around the Scalable Product Architecture (SPA), an €8B investment to lower costs for development and parts designed to be shared by all Volvo models. Model differentiation will be left for individual details important to customers. This reduces time even further because each system can be engineered independently — effectively decoupled from vehicle development — which reduces integration time. It also allows re-use of existing parts and systems reducing lead time by using existing tooling and processes.
Defined interfaces between modules and systems are vital to making this work. The analogy Carlsson used is to compare the method to Lego blocks, with interchangeable modules of varying functionality (and cost) to define a particular car’s value. He also stressed that the modular approach is important to future upgrades, in keeping with Volvo's rapid-follower business model. Individual modules can be improved to meet competitive pressures without waiting for a new platform, according to him.
A good example of this Volvo modularity is in their new engine strategy. The Volvo Engine Architecture (VEA) is composed of only 4-cylinder engines in eight or nine different end-products. VEA will replace eight different engine architectures, reducing complexity with up to 75% part commonality between individual engines. This too is not a new concept, but one that the industry has been striving to implement with varying levels of success.
Tooling lead times for key structural parts is a particular challenge. For certain parts, there remain physical constraints in fabrication. When these are coupled with late engineering changes in the program, tooling time can last longer than 20 months. In response, Volvo has developed strict rules for design maturity at Final Engineering Release, when part designs are formally ready to be tooled up.
“Engineers need to think of Final Engineering Release as truly final,” Carlsson said.
Continue reading »