Data science under pressure: From emergency to risk-awareness with nowcasting

157 Aufrufe
Published
What's up at our next August Pie & AI Suisse event? - From a corona emergency to risk-aware security operations with nowcasting, simulation, and geographic resource optimization.

Abstract:

The German federal health care system proved to be relatively elastic in building up emergency resources for COVID-19 patients. The capacity of intensive care beds to cope with the predicted epidemic peak load was increased from approximately 28,000 to 40,000 within a few weeks. This decision was made under uncertainty in February when infection numbers in Italy rose rapidly. Surprisingly, most of those additional COVID-19-capacities remained unused. The opportunity costs of this capacity expansion were paid for by potential patients who urgently needed hospitalization for other diseases and treatments. Yet, after easing the shutdown, who can rule out with certainty at least local resurgence of the epidemic? This presentation will focus on the return from an emergency operation to "risk-aware normal or safety operation", to which we will probably have to be prepared for the next 1-2 years. A solution approach for operative risk simulations based on geographic nowcasting, resource optimization is presented.

Short Bio:

Ulrich Reincke is Principal Data Scientist at SAS Institute in Heidelberg. For more than 25 years, he has been responsible for analytical decision support solutions. This usually involves better, more profitable or more transparent fact-based decision-making in automated, rule-compliant, auditable IT processes.

Project implementation experience:
-Machine learning for customer relation management: Customer acquisition, retention, cross-selling, up-selling, profiling, segmentation, lifetime value
-Fraud detection during the settlement of insurance claims,
-Insurance pricing and price optimization,
-Risk modelling for banks, insurances, utilities and public sector,
-Textual analysis of accident, claims or maintenance reports to identify relevant problems,
-Prediction of machine, system or asset failures,
-Intraday load forecasting based on smart meter data to manage balancing energy cost efficiently
-Supply chain optimization

Before joining SAS, Mr. Reincke held positions in the Information Product Department of Deutsche Börse in Frankfurt and in the International Trade Division of the World Bank in Washington DC. His academic background includes degrees in mathematics and economics as well as in international relations.

Sponsored by SAS

+++ deeplearning.ai ++++++++++++++
This meetup is in collaboration with deeplearning.ai. Please fill out the Google form (https://bit.ly/3bydbBi) to complete registration if you are interested in receiving course promo codes/AI news from deeplearning.ai.

https://deeplearning.ai
Kategorien
Corona Virus aktuelle Videos
Kommentare deaktiviert.