Company Cooperation Project

Inspired by data analyses in motorsports, a team of PwC data scientists and data engineers came up with the idea to build a slot car racing track equipped with several sensors to show data analytics possibilities at exhibitions. The key aspects of their project were:

– Creating an image recognition system that is able to identify the two race cars in a video stream captured by a camera placed next to the track.
– Developing new ways of controlling and monitoring the cars.
– Splitting the existing project backend into modular, domain-specific backends.
– Creating a central communication service that is responsible for information exchange between the frontend, the backends, and the cloud.

Sentyre is a German start-up that develops new technologies for fleet management. Our cloud solutions utilise German-developed artificial intelligence to demonstrate their potential in direct contact with the road. We analyse data in real time and process it in order to predict and avoid problems with tyres and rims. This includes air pressure that is set too low, which is the main reason for increased abrasion and fuel consumption. Sentyre collects data from trucks driving and develops a smart system to predict the best tyre for a certain truck and position on the truck. Students used data science algorithms and data visualisation to come up with a prediction model on the best tyres for certain trucks.

For millennia, humans are shaping the face of the world at an exponentially accelerating rate of change, especially in the last century. Man-made intervention to the vegetation around the globe is disturbing a natural equilibrium that existed long before the dawn of mankind with unknown consequences.

We want to mitigate this intervention, preserve intact vegetation wherever possible and get a better understanding of how the tourism industry contributed to the problems we observe nowadays in general, e.g., through deforestation for tourism facilities or spillage of ground water.

Navigating financial markets is prone to an overload of information, resulting in the challenge to make the right decision at the right time. This applies to the asset class of digital assets even more. Therefore, the Frankfurt School students enhanced the decision-making process by developing

1. a forecasting algorithm for predicting the returns of single digital assets using random forest algorithms and neural networks, and
2. by constructing a sentiment indicator for major cryptocurrencies by analysing tweets with the help of NLP algorithms.

Reliable forecasts and robust monitoring are vital for energy distributors to manage peak demand effectively, ensuring grid stability and meeting customer needs, with a focus on detecting external factors influencing demand anomalies. Students developed a Machine Learning Pipeline to detect outliers and forecast utility consumption, showcased through a data-driven dashboard with essential KPIs for easy maintenance. Steps included data pre-processing, shock detection with peak categorization and explanations, model training and testing, and automated demand forecasting with dashboard integration.

Students worked on a market trends project to understand market sentiment and investor activity across Xetra and Eurex. Project topics included clustering data, predictive power and analysing relationships between datasets. They used data science and machine learning to assess predictability and ultimately visualise the data with which a report was produced.