Weath­er dis­asters, pan­dem­ics & co: data plat­form for crit­ic­al haz­ard situ­ations un­der con­struc­tion

Natural disasters, pandemics, mass accidents: Time and again, situations arise in which people have to act quickly. The most recent example is the flooding in southern Germany. For experts, large amounts of data - sometimes highly complex - play a key role in the decision-making process. However, even experts are not always able to make logical decisions because they are often accompanied by considerable uncertainty factors. The EU project "Critical Action Planning over Extreme-Scale Data" (CREXDATA) is developing a platform for managing such situations. Based on data streams of extreme size and complexity, it will provide real-time analyses and predictions on how critical situations may develop. The project, consisting of 15 partner institutions, will run for three years and is being funded by the European Commission to the tune of around 8.7 million euros. Prof Dr Iris Gräßler's Chair of Product Creation at Paderborn University's Heinz Nixdorf Institute is involved in the project. The Paderborn scientists are primarily responsible for the design, coordination and testing of the overall system as well as for strategy development in the overall network. The project at the Heinz Nixdorf Institute is being funded with around 660,000 euros.

Many factors cannot be predicted

The flooding situation in southern Germany highlights the sad relevance of the multi-million euro project: "Weather services had predicted high levels of precipitation within a very short space of time. But what we don't know exactly: Which path does the water take on the ground? Where does it collect? What effect does it have on buildings, bridges, vehicles - and above all people? Many factors simply could not be predicted," says Dr Jens Pottebaum, co-initiator of CREXDATA and senior engineer at the Chair of Product Development.

Rescue robots and social media provide valuable data

During flood disasters, for example, rescue robots on the ground and in the air can send images and videos, while weather stations and satellites provide up-to-date data and forecasts. "With CREXDATA, we want to support the targeted use of these robots, among other things. This is because the extreme amounts of technical data have to be collated and analysed. Fire brigade operations managers have to interpret this information in order to make decisions," continues Pottebaum. Events and uncertainties are to be identified and subsequently visualised with the help of CREXDATA. Textual data such as tweets and posts from social media also provide valuable information: They use statements from the population to prove whether forecasts have actually materialised.

The project will run until the end of next year. The Technical University of Crete is leading the project.

For further information visit https://crexdata.eu/