Connected vehicles are extremely popular. This is especially true for the Chinese market. One of the automotive OEMs (Original Equipment Manufacturer) in the Chinese market decided to initiate a project that would help the company better understand the user behavior of its drivers. The results should be included in the development of new models and the car manufacturer specified a tight time frame for the development of this new service: six weeks from the end of March to the middle of May – during the global coronavirus pandemic. With the rollout of the solution, the automotive manufacturer can now optimally use the functionality of its connected cars to better understand its customers. New services can be rolled out based on the anonymous insights. At the same time, the optimized driving experience is a clear differentiator for the car manufacturer on the Chinese market. The company is establishing itself even more as a highly innovative car maker by providing its users with the latest functions and services.
With the analytics solution, our client is raising the driver experience of his customers to a new level and is gaining valuable insights for the development of future vehicle generations.
Xu Zhao, Account Manager, T-Systems China
Connected car – that doesn‘t just mean connectivity. It is also big data: operating sensors in the cars continuously record and store operating data. Every day, immense amounts of data from various sources and in different formats are generated in the backend. For the respective OEM this connected vehicle backend is operated by T-Systems in a private cloud in China. The evaluation of this data is of great interest for the brand in order to offer its customers excellent service (including new in-car offers), increased driving safety and an excellent driver experience. However, the evaluation of data from various sources is a field with its very own specific challenges. This not only includes the preparation of the data, but also the ability to provide actionable business insights based on the analyses.
Since T-Systems not only operates the connected car backend, but also has great big data expertise, the automotive manufacturer commissioned the IT service provider to develop the analytics solution. The extensive infrastructure resources for modeling the solution and machine learning come from the Huawei Cloud.
Within a few days, T-Systems established an agile team and delivered the solution for the department as a web service by the end of April. The business department units can access the service via browser. The six-week project during the coronavirus pandemic included a final two-week user acceptance test. First, the T-Systems team extracted the relevant data from the connected vehicle backend in T-Systems‘ private cloud. The data was largely structured and also had a similar data format. Nevertheless, the data records had to be cleaned up prior to transfer (via rvs, the computer system from T-Systems) to the public cloud. The team led by project manager Xiaojie Weng transferred the cleaned data sets to the Huawei Cloud in order to use the capacities there for the rapid development of the analysis service. Among other things, the Hadoop PaaS provided in the public cloud was used by the team. The team also set up the web frontend, which the OEM‘s employees can use to access the anonymous data analysis via a web browser.
