KIRAS Security Research

Call results by calendar year

Call results by funding instruments

Projects of the joint German-Austrian call

Cooperative Research and Innovation Projects >Call 2018 >

PrEMI - Predictive Analytics for Emergency Call Infrastructure

Data analytics and predictive methods for optimizing the operation of critical communication infrastructures, required for emergency call handling in public safety answering points (PSAP).

Ensuring the best possible assistance in emergencies and hazardous situations starts with the reliable and efficient receiving and handling of emergency calls at control centers like public safety answering points (PSAP). During the police assistance process, it is often a matter of seconds to be at the scene in time in order to prevent crime or bring the perpetrators to justice. 

The communication infrastructure data generated in the control center process – in the case of police emergency calls, these are metadata of the emergency calls themselves and the operating data of the devices that are generated during the ongoing monitoring of the communication infrastructure – contain valuable information indicating possible acute or evolving problems. The analysis of these data, the extraction of patterns and the intelligent use of the extracted information via prediction models provide a great potential for the improvement and optimization of emergency call handling. 

The overall objective of the project PrEMI is the reliable and quality-assured operation of the critical communication infrastructures, required for emergency call handling (police and European emergency telephone number) in a PSAP. Algorithms and models are to be developed that provide information and predictions while the system is in operation using real-time data to improve availability, reducing response and processing times as well as downtimes. For this purpose, the operating data of the emergency call infrastructure and the anonymized metadata of the emergency calls (Call Data Records) are merged, prepared, analyzed and prediction models derived. 

From a technical and scientific point of view, the project aims at investigating and developing suitable methods for data analysis and optimization: 

  • Statistical reliability and time series models for the purpose of root cause analysis (Predictive Modeling).
  • Deep-Learning algorithms for pattern recognition and for prediction of problems and downtimes.
  • Application of Automated Machine Learning for selecting the best possible machine learnig or prediction method resp. parametrization therefore.
  • Optimization methods for improved planning of resources, maintenance and prioritization of emergency calls.
  • Sociological study of the relationships between the quality and the reliability of the emergency call process and the emergency call behavior of the persons seeking help.

The result of the project will be demonstration software for selected methods and algorithms which provides proof of concept on a laboratory scale, based on test data of selected PSAPs. 

The project will show which methods of data analysis and optimization have the greatest potential for practical use in the control center process in order to reduce processing times as well as downtimes, and optimize availability. 

Project management 
DI Ulrike Kleb
JOANNEUM RESEARCH Forschungsgesellschaft mbH, POLICIES

Project partner 
Bundesministerium für Inneres (BMI)
Technische Universität Wien, Institute for Logic and Computation
Universität Graz, Institut für Soziologie
Axtesys GmbH
NTT Austria GmbH

Contact 
DI Ulrike Kleb
JOANNEUM RESEARCH Forschungsgesellschaft mbH
POLICIES – Institut für Wirtschafts- und Innovationsforschung
Data Analytics and Statistical Modelling

Leonhardstraße 59, 8010 Graz, Austria

phone: +43 316 876-1555
e-mail: ulrike.kleb@joanneum.at
web: www.joanneum.at/policies