Aim of the project is to measure training load, to predict effects of training stimuli and to optimize training management based on individual physiological reactions of athletes. In cooperation with coaches and 40 top athletes of the Austrian Rowing Federation, the research network Data Science and the Institute of Sports Science of the University of Vienna will work together in the fields of statistics, computer science, mathematics, sports science and sports medicine.

Information on training content and intensity, parameters of performance diagnostics and various biomarkers will be processed using state-of-the-art machine learning approaches. The goal is to quantify internal load and predict the effect of a specific training stimulus on individual athletes. A further objective is to gain better understandings on which parameters serve best as indicators for these calculations and how modern technologies (e.g. wearables) can be optimally used to collect the required data.

The group of sports science provides the sensor technology for rowing ergometers and boats. Additionally, it will focus on the implementation of the "Performance Potential Double Model" (PerPot DoMo) to deliver orientation values to the developers of the machine learning model.

The final product should be an app that collects the athletes' load, health and recovery states and uses this information to provide relevant feedback for optimizing and individualizing training management. The project is carried out as a pilot in the field of rowing. In a later stage, the insights gained can be transferred to disciplines with similar training requirements, such as swimming, triathlon, cycling, cross-country skiing and long/mid-distance running.

The project started in July 2022 and will last for 3 years. More information is available at https://airow.univie.ac.at/.