FEV is a globally leading engineering provider in the automotive industry and an internationally recognised leader of innovation across different sectors and industries. Professor Franz Pischinger laid the foundations in 1978 by combining his background in academia and engineering with a great vision for continual progress.
The company has supplied solutions and strategy consulting to the world’s largest automotive OEMs and has supported customers through the entire transportation and mobility ecosystem. Besides the automotive industry, FEV applies its forward thinking to the energy sector, and its software and system know-how is leading the way to make intelligent solutions available to everyone. FEV brings together the brightest minds from different backgrounds and specialties to find new solutions for both current and future challenges, working with over 7,300 highly qualified employees at more than 40 locations globally.
In India, FEV has a strong presence since 2006 and is located near the company’s clients at six major mobility and industrial hubs: Pune, Chennai, Delhi, Jaipur, and Bengaluru. With more than 960 engineering experts, FEV India operates state-of-the-art workshops and test facilities.
Engineering data analytics
On the booming journey of digital transformation, data is the new fuel, and engineering analytics is translating it into a green fuel for the automotive industry. Engineering analytics is a multifaceted field that combines physics-rich models, statistics and mathematical principles, and analytical techniques to gain valuable insights and cognitive decisions which curate research-based innovative methods more effectively in engineering applications. It involves the use of data analysis, statistical modeling, optimisation methods, physics simulations, signals, and systems to solve complex engineering problems and improve overall system performance. Due to ongoing digitalisation of the automotive ecosystem and vehicles that are highly connected, driven by software and packed with sensors, a large amount of data is continuously generated which can be leveraged for more efficient and optimised operations of the entire value chain. Engineering analytics exemplifies the principle of optimising the value chain, thereby transforming innovative ideas into solutions that establish new industry benchmarks.
Analytics for the automotive industry
The automotive industry is currently experiencing a remarkable transformation, marked by an abundance of data derived from extensive testing endeavors such as vehicle-on[1]road operations, opening previously unfathomable potential. This enormous influx of data has immense possibilities for the automotive sector. Engineering analytics can effectively leverage data-driven analytics applications for optimising vehicles as well as for the benefit of society.
Vehicle optimisation
Product development: Automotive businesses can strengthen their products & services by designing and engineering vehicles that meet dynamic market expectations by statistically analysing client preferences, usage patterns, and customer feedback.
Operational efficiency: Engineering analytics provides improvements in production processes, preventive maintenance, and supply chain management, which result in lower costs and higher productivity.
Personalised customer experiences: Individualised services, features on demand, customised marketing initiatives, and improved post-purchase support by utilising data on consumer behaviours and preferences increase customer satisfaction and loyalty.
Benefiting the society
Enhanced road safety: The role of highly sophisticated ADAS applications based on multi-sensor systems is to minimise the number and severity of road accidents. Application of intelligent systems based on data-based models alerts the driver in case of probable collision, it also enforces following the traffic regulatory system. These advanced systems have become integral to modern vehicles, enabling a host of robust functionalities aimed at mitigating accidents and promoting responsible driving practices. These ADAS components eventually mature towards autonomous driving systems.
Environmental sustainability: Enhancements based on guided learning in vehicle design and ever evolving analytics services for more efficient vehicle operations, conclude in lowering the carbon footprint and optimised usage of natural resources – materials and energy.
Smarter urban planning: Automotive smart urban planners can design and optimise city layouts, ease traffic, and enhance overall urban mobility by considering analytical data on traffic patterns, commuter preferences, and transit requirements.
Growing demands & challenges
With technological advancements in the automotive industry, there is growing demand for Engineering Data Analytics. The spectrum is covering everything from big data solutions for connected vehicles & ADAS/AD, prognostic and health analytics, digital twins for road & traffic scenarios, EV battery & vehicle, and an unbelievably large set of point applications like dynamic vehicle mass estimation, fleet analytics & scheduling to driver behaviour analytics, infotainment system V&V, cognitive QA, etc. While there are myriad applications and use cases, challenges to it are equally numerous. Some of the critical considerations are:
- Integration of heterogeneous platforms and disparate data sources.
- Reference data generation for ADAS/AD features development.
- Coverage of edge cases for the ultimate safety and security of applications.
- Data management & controlled distribution.
- Transparent/trustable & ethical intelligence for critical analytics. applications.
- On-edge and cloud analytics.
- Customised signal processing algorithms designed for on-edge analytics.
- Cost constraints and affordability.
- Scalability & extensibility across types and usage of vehicles.
- Availability of multi-disciplinary but niche skills & expertise.
FEV’s readiness
The challenges for the application of analytics in the automotive industry are extensive. Considering the vast number of challenges, addressing each one individually is thus quite unrealistic. However, modern techniques like digital twins, ML & AI, and hybrid modeling enables addressing some of these challenges together. Furthermore, it requires an almost 180-degree shift in solution design – a transformation from data product as a service to platform as a service and analytics as a service.
Such a platform enables the extensibility of fit-for-use components ready to deploy for varied requirements, offering services like feature on demand, and a lot more derived benefits. FEV has proudly held the esteemed position of a global industry leader, steadfastly demonstrating unparalleled expertise and skills. Since its establishment in 1978, FEV has consistently remained at the forefront, blazing a trail of excellence in Engineering Analytics that has earned global recognition and acclaim.
Illustrative platform solutions
Automatic data annotation and labeling framework – Intelligent way to generate high-quality labeled data for ADAS/AD features development, indigenously developed cloud platform for creation, visualisation, and review of labeled data, state-of-the-art AI models and customised algorithms offering excelling performance, built-in model retraining adaptive for changing geographies and data.
Prognostic and health analytics platform – Enabler for new services and next-generation products in automotive asset lifecycle management transforming scheduled maintenance to a predictive, hybrid approach of data-driven modeling augmented with physics-based modeling & advanced analytical models (digital twin).
Powertrain data analytics platform – A heterogeneous platform integrating domain-driven knowledge with data-driven models for accurate analytics, collection of reusable assets, and components for reconfiguration of analytics services based on requirements. Some of the services realised are driver skills & behavior analytics, vehicle operations KPIs tracking, components health scorecards, etc.
Intelligent automotive test scheduling – In the fast-paced world of automotive, testing accuracy and efficiency are paramount. To streamline this, the FEV analytics framework leverages the power of Cognitive AI to optimise the test job scheduling and enhance productivity. By employing reinforcement learning algorithms, this framework learns from real-time data to make cognitive decisions. It considers various factors such as test priorities, resource allocation, utilisation, and minimum time consumption to execute an optimal job schedule. This minimises idle time, maximises resource utilisation, and significantly reduces overall testing duration. The framework unlocks new possibilities for efficient testing and faster product development, which leads to enhanced quality and customer satisfaction in the automotive industry.
Apart from customised analytics platform solutions, FEV is delivering Analytics as a Service for application specific requirements and its deployment at an industrial scale.