Driven by AI: How we develop AI-featured solutions at Porsche Digital
At Porsche Digital, AI is a core technology and allows to automate and optimize products and business processes in many different areas. The way the AI team @ Porsche Digital works, enables to develop AI based solutions for internal and external customers at a high velocity.
How? Peter Wolf, Data Science Manager at Porsche Digital, explained it at the world’s largest developer event, the WeAreDevelopers World Congress in Berlin.
Digital innovations and technological progress are key drivers for the future of the company at Porsche. Porsche Digital is a key partner for the digital transformation of Porsche and state-of-the art technology is at the core of its products and services.
The AI Team @ Porsche Digital tackles a variety of data-driven challenges in several fields of the company. For example, in Research & Development it helps in the fields of aerodynamics, in-car climatization and component tests. In production, AI-based approaches were applied in logistics, disposition, and production planning. Algorithms also analyze Vehicle field data to gain insights on the usage or the battery.
Innovation driven by hypotheses, data and of course the customer
No matter in which field: the innovation process is hypotheses-, data- and customer-driven. It moves through five phases, and workflow ensures that along the process and with increasing investment the risks decline. For each phase there are questions, that help the team to determine on how to drive innovation forward. To develop a productive AI-based solution starting from an idea, standard agile (i.e., iterative and incremental) approaches are followed.
- Phase 1: Customer-Problem Fit
Is there a problem worth solving? Are we able to propose a solution that is desired?
- Phase 2: Problem-Solution Fit
Do we believe we will be able to create a viable business model?
- Phase 3: Product-Market Fit
Can we build this technically? How can we keep learning from the customer?
- Phase 4: Pivot
Are we on the right track? Will the value be higher if we change the concept?
- Phase 5: Growth
Will we be able to scale up to generate a profit?
When is Machine Learning the right approach?
Peter and his colleagues focus on the application of Machine Learning, specifically Deep Learning based on (deep-) neural networks to create algorithms that learn from data. These approaches help, if data exists that covers patterns required to solve a problem and if manually defined rules or models get too complicated to capture a problem at an adequate level of abstraction needed for a solution.
One example of what the AI team @ Porsche Digital has developed so far is Sounce. It enables automatic detection of unwanted noise in real-time and has been mainly designed for production and development processes. Exemplary use cases are component endurance runs or end-of-line test benches for drive systems, closing systems or imbalance testing. The Software-as-a-Service solution was developed together with colleagues from the development department of Porsche AG.
Sounce is a great example, because it shows how their focus, tech stack and way of working enable the AI team @ Porsche Digital to develop state-of-the art AI-based solutions. It’s how the international team creates innovation for Porsche and beyond.
About this publication: Where innovation meets tradition. There’s more to Porsche than sports cars — we are developing new digital products and services — always with our customers in focus. On our Medium blog, we tell these stories. It’s about our #nextvisions, emerging technologies, and the people that drive our digital journey. If you want to know more, follow us on Twitter, Instagram and LinkedIn.
