Performance study of crowd flow in academic buildings of an institution

An institution’s academic buildings are an important meeting point for students, and the crowd flow affects its ecological parameters. Therefore, it is necessary to measure the ecological parameters and crowd flow in such a building. Tracking the density of large-scale people and environmental param...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:CCF transactions on pervasive computing and interaction (Online) 2023-12, Vol.5 (4), p.367-381
Hauptverfasser: Chellaswamy, C., Rao, C. Subba, Geetha, T. S.
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:An institution’s academic buildings are an important meeting point for students, and the crowd flow affects its ecological parameters. Therefore, it is necessary to measure the ecological parameters and crowd flow in such a building. Tracking the density of large-scale people and environmental parameters is a challenging task. In this study, a lightweight smartphone-based crowd flow measurement (SP-SP-CFM) is introduced for tracking the crowd flow in academic buildings. The proposed SP-CFM searches the requests enabled by the user’s smartphone at regular intervals. SP-CFM has been implemented, and a different test has been carried out in the academic building in real-world settings. The SP-CFM was deployed in different locations, including the evacuation passage on the first floor and two classrooms. The ecological parameters, such as temperature and CO 2 concentration, have been studied at different locations in the evacuation passage. The effect of people’s walking behaviour has been studied in four different modes of smartphone operation. The performance of the suggested SP-CFM is evaluated with different models of smartphones with variable walking speeds. The results show the proposed SP-CFM provides a tracking accuracy of 93.6% in the Wi-Fi registered mode.
ISSN:2524-521X
2524-5228
DOI:10.1007/s42486-023-00134-9