Tomorrow’s cars will demand software and product engineering combined, says Derek Piette, product manager for application lifecycle management and product lifecycle management integration, at Siemens PLM Software.
Software is increasingly part of our daily lives – it’s embedded in the products we use every day. Modern cars have 100 times as much code as the F-22 fighter jet, and that number keeps going up thanks to connected in-car services and the move to autonomous and hybrid and electric vehicles.
A decade ago, software accounted for less than 20% of the cost of a vehicle and did little to differentiate one model from another. Now 80% of all product innovation and differentiation comes from the combination of software, electronics and electrical systems, according to Daimler.
To bring together the different application and product lifecycle management worlds, software development tools need to be tightly integrated into the product process. Here they need to let developers code in an environment they can collaborate in and be productive in.
The interactions between software and other product functionality means that smart products can have far more complex defects. For example, imagine having new navigation software installed on a car only to find that this update has disabled the button on the vehicle dashboard that opens the garage door.
Software problems go well beyond simple annoyances though. Vehicle recalls and warranty claims are expensive and they’re on the rise – in 2016, a record 927 recall campaigns took 53.2 million vehicles off the road in the USA alone with software issues increasingly to blame. Problems range from a security flaw in an entertainment system to software bugs that caused the doors to open unexpectedly.
Before 2011, it was rare for 5% of vehicle recalls to be caused by software issues; by 2015, almost 15% were software related. The figures are even worse in other industries – in 2014, 24% of medical device recalls were because of software.
A decade ago one could get away with designing a product and then developing the software to power it because it was an isolated component. Now that electromechanical systems are integrated and interconnected this is no longer sustainable; integration needs to start at the very beginning of a design.
Integration means understanding the relationship between different modules. Change requests during the product development cycle must now be analyzed for their impact on both software and hardware components, and the implementation of those changes needs to be coordinated between teams to streamline operations. Minor differences in hardware can significantly affect software performance, so one needs to be able to ensure the right balance of software and hardware.
Traceability across different product systems – from the source data for all the different variants – is key to managing dependencies between systems, producing accurate bills of materials to build from, and to track and understand why decisions were taken and any problems that arose.
As we move to a world of mass customization, driven by the shift toward additive manufacturing processes like 3D printing, tracking the exact combination of hardware and software components in a product will be vital for service and support. Rapid prototyping will become the norm, shortening the hardware development cycle. These trends will make integrating software and hardware development with tools like digital twins even more important.
A digital twin of the physical components that integrates with software management tools will give a living product version that can make manufacturing and support far more flexible and responsive, but it will only work if the correct collaborative tools are in place.
If one is not thinking that far ahead, integrating software and hardware development can have a significant impact on the bottom line very quickly. By knowing which hardware and software components are used in which variants of specific vehicle programs in different territories around the world, down to the individual vehicle identification number, Ford was able to update electronic control units with software issues in the field instead of replacing the hardware. Over three years, that saved the company US$100m in warranty costs – a great payoff for a system that also increases customer satisfaction.