Measuring gait space: new dimensions for crowd flow

Peter Thompson

doi:10.1016/j.firesaf.2026.104740

Measuring gait space: new dimensions for crowd flow

2026-05-23

0

PRE

AI

DOI：10.1016/j.firesaf.2026.104740

Original

Original request for help

Save

Abstract

En 中文

Widely used microscopic crowd movement models often rely on simplified movement data, or calibrated 'social force' approaches to simulate evacuations, without including validated data or processes to model lateral movements. This often necessitates limiting flows with basic, long-established 'macroscopic' data. This study begins to address this gap by processing pedestrian trajectories for different corridor width experiments (0.9-1.2 m) to derive step events, vectors, relative lateral relationships, and forward speeds. Lateral-offset frequency distributions, inter-personal spacing, and individual velocities are calculated. Two distinct walking modes are defined: 'direct ahead' and 'fully offset' gait conditions. The lateral offset distances significantly influence headway, indicating that interpersonal spacing is inherently multi-dimensional, as pedestrians seek to achieve the required 'gait space' for safe, comfortable walking. Even in narrow corridors (0.9 m), pedestrians strongly prefer laterally-offset walking to avoid leg movement impediments, sometimes twisting their torsos to maintain this mode in close proximity to a wall. We highlight the need for crowd movement models to incorporate 'gait space' parameters, including discrete walking modes, lateral positioning relationships, and additional demographic considerations. The use of lateral 'gait space' units of 0.35 m for constrained geometries is suggested for future simulations and fire safety exit design.

Keywords:

Egress

Human factors

Gait space

Crowd flow

Lateral offset

Evacuation modelling

Pedestrian dynamics

Trajectory analysis

Evacuation route design

Measuring gait space: new dimensions for crowd flow

Abstract

Journal

Organization