Artificial Neural Networks Predicting Pedestrian Dynamics in Complex Buildings

Tordeux, Antoine; Schadschneider, Andreas; Seyfried, Armin; Chraibi, Mohcine

doi:10.1007/978-3-030-28665-1_27

Contribution to a conference proceedings/Contribution to a book

FZJ-2019-05548

Artificial Neural Networks Predicting Pedestrian Dynamics in Complex Buildings

Tordeux, A. (Corresponding author) ; Chraibi, M.FZJ* ; Seyfried, A.FZJ* ; Schadschneider, A.

2019
Springer International Publishing Cham
ISBN: 978-3-030-28664-4 (print), 978-3-030-28665-1 (electronic)

Steland, Ansgar (Editor) Chapter 27 ; ISSN: 2194-1009=2194-1017 ; ISBN: 978-3-030-28664-4=978-3-030-28665-1
Stochastic Models, Statistics and Their Applications, SMSA 2019, Dresden, Germany, 6 Mar 2019 - 8 Mar 2019 Cham : Springer International Publishing, Springer Proceedings in Mathematics & Statistics 294, 363 - 372 (2019) [10.1007/978-3-030-28665-1_27]

This record in other databases:

Please use a persistent id in citations: http://hdl.handle.net/2128/23606 doi:10.1007/978-3-030-28665-1_27

Abstract: The prediction of pedestrian movements in complex buildings is a difficulttask. Recent experiments have shown that the behaviour of pedestrians tends todepend on the type of facility. For instance, flows at bottlenecks often exceed themaximal rates observed in straight corridors. This makes pedestrian behavioursgeometry-dependent. Yet the types of geometries are various, and their systematicidentification in complex buildings is not straightforward. Artificial neural networksare able to identify various types of patterns without supervision. They could be asuitable alternative for forecasts of pedestrian dynamics in complex architectures. Inthis paper, we test this assertion.We develop, train and test artificial neural networksfor the prediction of pedestrian speeds in corridor and bottleneck experiments. Theestimations are compared to those of an elementary speed-based model. The resultsshow that neural networks distinguish the flow characteristics for the two differenttypes of facilities and significantly improve the prediction of pedestrian speeds.

Contributing Institute(s):