KIT | KIT-Bibliothek | Impressum | Datenschutz

Decision Support for Road Safety: Development of Key Performance Indicators for Police Analysts

Meißner, Katherina; Rieck, Julia

In 2017, five out of 100,000 people were killed by road accidents in Europe. In order to reduce this number with appropriate measures, the police nowadays manually defines combinations of accident attributes (e. g., accidents on slippery road surfaces at night), which then form the basis for tracking the number of accidents over time. The aim of this paper is to combine the following data analysis approaches in order to detect interesting attribute combinations, also referred to as “itemsets”, relevant for current and future observations. The resulting combinations are proposed to the police as new key performance indicators and can also be used directly for planning police measures to increase road safety. A four-stage decision support system is introduced that employs frequent itemset mining in the first stage. The temporal aspect of traffic accident data is illustrated by time series containing, for each itemset, the relative frequencies of accidents with the corresponding attribute combination. In the second step, the time series are grouped according to their shape by time series clustering and classification. In the third step, we determine the optimal forecasting method for each generated cluster of time series. ... mehr

Verlagsausgabe §
DOI: 10.5445/KSP/1000098012/12
Veröffentlicht am 06.10.2021
Cover der Publikation
Zugehörige Institution(en) am KIT Institut für Wirtschaftsinformatik und Marketing (IISM)
Publikationstyp Zeitschriftenaufsatz
Publikationsjahr 2020
Sprache Englisch
Identifikator ISSN: 2363-9881
KITopen-ID: 1000138752
Erschienen in Archives of Data Science, Series A
Band 6
Heft 2
Seiten P12, 18 S. online
KIT – Die Forschungsuniversität in der Helmholtz-Gemeinschaft
KITopen Landing Page