Optimal Measurement Interval for Pedestrian Traffic Flow Modeling

Abstract

Measurement interval is crucial in obtaining stable traffic flow parameters, such as traffic flow rate, density, and speed. Credible relationships among these parameters and reasonable capacity for analysis are also dependent on the appropriate measurement interval. Nevertheless, the measurement interval has not been mentioned explicitly in most previous studies related to pedestrian traffic flow. Different researchers adopted various time intervals for traffic flow parameters measurement, and the reasons for the choice of a certain measurement interval were usually not clearly stated. This paper deals with the impact of different intervals on the results of pedestrian traffic flow parameters counting and the methods to determine the appropriate measurement interval. It takes pedestrian traffic in Shanghai Metro stations into an observation context. By video recording, pedestrian traffic parameters, including volume and density, are extracted manually through recording tapes using an initial 5-s interval for level passageway, ascending stairway, descending stairway, and two-way stairway. Taking 5 s as the basic incremental unit, traffic parameters for 10-s interval, 15-s interval, and other analogical time interval values are acquired in sequence. On the basis of the principles of measurement interval selection, the appropriate measurement interval is determined by comparing its influence on stability of hourly flow rate and invariability of pedestrian traffic status when applying different time interval values. It is revealed that a 30-s interval value is optimal for all the observed facilities in this research context. From the analysis of optimal measurement interval, it could also be inferred that traffic stability of level passageway is higher than that of stairways, and a one-way stairway is more stable than a two-way stairway. The collected data, proposed methods, and revealed pedestrian flow characteristics could be helpful in improving the theories on pedestrian traffic flow, and serve as valuable references for similar pedestrian context studies.