ANALYSIS OF PEDESTRIAN FLOW CHARACTERISTICS ON ESCALATORS AT METRO STATIONS

Abstract

India is a fast-growing nation with an extending economy, population, and urbanization resulting in space constraints in urban and semi-urban areas. A world-class transport infrastructure and an effective public transport system are required to manage the transportation needs of the sprawling urban population in cities and towns which can help in reducing traffic congestion conditions in the urban areas. Metro is one of the most efficient modes of public transport in urban areas to handle traffic. Stations are the locations in metro systems where passengers interact continuously to use the facilities and services. The stations can be classified as terminal stations, transfer stations, and intermediate stop stations. The passenger movement requirements at these stations differ accordingly. Many times the stations are located either underground or elevated above the ground level. Passengers need to change floors between entry/exit and platform levels, in general, or at transfer stations to use different connecting metro lines. Therefore, the stations shall be designed to handle large numbers of passengers causing minimum possibilities of delays, inconvenience, and accidents to them. To expand the passenger handling capacity of the stations and to allow for easy and convenient pedestrian movement, inclined pedestrian facilities like escalators are typically provided at the stations to move between floors. Escalators are mechanized pedestrian facility which requires a bit of experience to use it. This relates to stepping on the first step and stepping out of the step when it reaches the floor level. It may have an impact on the pedestrian flow on the escalators. The flow on the escalators at metro stations usually does not remain uniform or continuous in nature. It builds with the arrival of a train at a platform or with respect to the departure time of the train from the platform. The flow on the escalators remains for a short period and therefore it influences the time interval that shall be used for the extraction of the flow data or any other desired data. The examination of the literature on this aspect indicated that the researchers have extracted the pedestrian-related data using time intervals as small as 10s and as large as 1 hour. Out of these, there are very few studies that are on escalators, and these are not conclusive on the selection of time intervals for data extraction. This stimulated to take up a research work to explore the issue of the time interval for data extraction for pedestrian flows at escalators. Another issue observed in the escalator-related literature was the absence of sufficient material for analysis of pedestrian flow characteristics on escalators. The constant speed of escalators might have been the deterrent in developing relationships between different characteristics, but it shall not deter studying the flow, space, and density characteristics related to pedestrians while using the escalators. This will allow us to understand the flow patterns which may be useful in taking decisions to deal with flow problems during peak conditions. This aspect was taken up as a second aspect to study and examine. The limited guidelines on the pedestrian flow at escalators indicated that there usually remains a vast difference between the theoretical capacity and maximum flow in the field (operational conditions) at any site. The theoretical capacity is calculated based on step occupancy which is considered as two pedestrians per step to arrive at the maximum value. This condition mostly does not occur due to varied reasons like personal privacy, standing with an unknown person very close, bigger body size of a pedestrian occupying more lateral or longitudinal space as compared to step dimensions (thus not allowing two persons to stand side by side or one behind other), a pedestrian carrying baggage or a baby (again causing step vacancy), etc. This issue stimulated to take up two issues – one, the adequacy of step size in the light of body dimensions as stipulated in different guidelines around the world, and second, the mismatch between theoretical capacity and filed flow wherein the maximum field flow will never reach the theoretical capacity. The second issue has a significant impact on the decisions regarding when to upgrade the escalator facilities. While working with the escalator capacity the escalator use etiquette also needs to be studied and discussed. Escalator etiquette varies from country to country, and in certain places, it is not even observed. It may be stand only or stand on one side and walk on the other side on the escalator step. The second case influences the flow and capacity of the escalator. Considering this, a reference capacity can be estimated which can be used as a comparison value for maximum flow in field conditions (instead of the theoretical capacity). Finally, it is appropriate to define the operation levels so that the operational condition of the escalator can be ascertained at any point in time which may help in taking upgradation or other policy decisions to ease the pedestrian flows at metro stations. In light of the above discussion, the present research focuses on the analysis of pedestrian flow characteristics including capacity analysis and developing the Level of Service (LOS) of the escalators installed at metro stations. To address the objectives decided for this research, two metro services, Delhi metro in North India and Bangalore metro in South India, were selected. Four stations were selected on the Delhi metro and three stations on the Bangalore metro. The selection was done based on the volume of commuter traffic. Different escalators were identified at the selected metro stations. Data were collected through a videographic survey method. Considering the suggestions given in the literature, the pedestrian flow data are extracted for a successive time interval of 10s to get the instantaneous peak. This allowed to study the variation in pedestrian flow with respect to different extraction time intervals. To get pedestrian space or density the pedestrians on successive 10 steps of the escalator were considered. Extraction was done for every ten seconds interval. Other inventory data like the location of the escalator with respect to the platform/entry/exit of the station, frequency of trains at platforms, flow movements and patterns, gender division, age distribution, pedestrians carrying baggage or babies, etc., and pedestrians walking on the escalator were also recorded. Pedestrian walking speeds were also recorded. To satisfy the first objective i.e., arriving at the optimal time interval to extract flow data on escalators, the successive 10s-time interval flow data were used. The successive data were aggregated to get pedestrian flows in data extraction time intervals of 20s, 30s, 40s, 60s, 2min, 3min, and 5min. Average pedestrian flow, peak pedestrian flow, standard deviation, coefficient of variation, and standard error of mean were calculated for all time interval data. Pedestrian flows were plotted against the data extraction time intervals for all the above-mentioned categories. It was observed that as the data extraction time interval increases the peak flow value reduces. However, the statistically significant difference was not observed in the average flow values across all the time intervals considered. COV was found to decrease with an increase in the data extraction time interval. The variability in the statistical measures was found to be quite less in the data extracted for the time interval of 1 min or more. Further examination of flow attributes and parameters was done across different data extraction time intervals of up to 60s.A cumulative frequency distribution is plotted for the peak flow categories to get the 98th percentile value. This was correlated with the peak flows during different time intervals up to the 60s. It was further observed that the peak flow values fall within a short range of 20s and 40s. The 98th percentile instantaneous peak pedestrian flow was found to occur around the 30s data extraction time interval. Then by using the 30th Highest Hourly Volume concept related to road (vehicular) traffic, a check is being made to see how well the 98th percentile value fits in the rank ordering of the peak pedestrian flows observed during the total period of the data recording. It was found to be related to the 5th rank order. This satisfied the first objective. The pedestrian flow characteristics on the escalators were analysed to add to the existing knowledge on the operation and use of the escalators at metro stations. It was observed that the escalators carrying flow from a platform after the train had arrived reached a higher instantaneous peak than those carrying flow to a platform. This can be attributed to the platoon effect. The flow values during evening periods were found to be higher as compared to flow values during morning periods. These values were higher than the flow values observed at metro stations in China or England or on escalators at railway stations. The same inference was made for pedestrian density. The most significant relationship between flow-density has been found as quadratic and between flow-space as logarithmic. The flow has been found to increase in proportion to the density of up to 2.5 pass/m2. After 3.5 pass/m2, the curve starts flattening. This The quantitative LOS criteria and threshold values for the escalators at metro stations were developed based on flow rate, peak flow to capacity ratio (for stand only and stand walk case), step occupancy, and minimum space at maximum flow characteristics. The LOS thresholds are developed by using clustering techniques. Six LOS categories were defined as A to F. Pedestrian space less than 0.18 m2, flow rate of more than 148 ped/m/min, step occupancy of more than 1.58 ped/step, and volume to capacity ration of more than 0.76 were classified as LOS-F. The examination of escalators at Delhi metro stations indicated that they were operating at LOS-D to LOS-E whereas escalators at Bangalore metro stations were operating at LOS-B to LOS-D. This research offers a detailed understanding of the flow characteristics of escalators installed at metro stations. The research has provided directions regarding the optimal data extraction time interval which can be used in studies related to pedestrian flows at escalators. The same can be used for data extraction in other pedestrian facilities too. The analysis or pedestrian flow characteristics provides insights into the decisions the pedestrians take under peak flow conditions and provides the possible relationships between the flow characteristics which can be considered for escalators. The flow conditions in India are found to be much more constrained as compared to the flow conditions in other countries. The suggestion to make changes in the step dimensions may bring in further changes in the flow characteristics and will be required to be studied again after such implementation. The reference capacity approach can be termed as revolutionary in flow management using escalators. So far, the use of theoretical capacity restricts the practitioners to have qualitative measures in hand to take judicious decisions. Finally, the development of LOS criteria will further help the practitioners in the flow management at metro stations. The research has thus contributed to the field of pedestrian flows and facilities, and management of flows in marked areas.