MODELLING OF PEDESTRIAN BEHAVIOUR IN URBAN AREAS

Abstract

Walking is the most common mode of transport for majority of the people in developing countries. Rapid motorization in the last few decades has led to widespread problems to non-motorized users in India. Most of the cities provide alarge share of the scarce road space to motor vehicles which contribute to minor percentage of trips made, whereas, the public transport, pedestrians and cyclists are forced to have a smaller share. The non availability of exclusive pedestrian facilities forces pedestrians to walk on the carriageway. Adetailed review of earlier research indicated that majority of the studies are reported for exclusive pedestrian facilities and the influence of frictions like vehicular traffic, parked vehicles, obstructions and barriers on pedestrians walking on the carriageway are not studied. Further, the majority of these studies are for developed countries and the complexity of Indian cities is such that the standards developed for these countries cannot be directly applied to India. The following specific problems of pedestrians need to be addressed i. The land use patterns of Indian cities are complex and the pedestrians walking on the different land uses are to be analyzed to find the effect ofdifferent land uses on pedestrians. This will help in design of facilities catering to specific walk purposes like shopping, commuting, etc. ii. The speed analysis of pedestrians on exclusive facilities is done at macro level in most of the reported studies. A micro level assessment of pedestrian speed considering various combinations of factors influencing them will highlight the combined effect of various factors on their speed. in iii. The behaviour of pedestrians on the carriageway under mixed traffic condition has not been studied in detail so far. The interaction of pedestrianswith vehiculartraffic and frictions like parked vehicles and obstructions on the carriageway must be studied to fully understand the influence of these factors on walking patterns and behaviour, iv. Most of the Level of Service models reported in literature are based on the pedestrian flow and area module. The safety and convenience of pedestrians is not fully taken into consideration. This is of utmost importance because the pedestrians walking on exclusive facilities are affected by the presence of frictions like surface irregularities, bottlenecks, hawkers, vendors and obstructions. The analysis of these factors will aid in enhancing the knowledge of pedestrian behaviour under varying conditions and their Level of Service (LOS) In order to address these problems, pedestrians from 19 study locations across 5 five Indian cities are considered and their walking speed, flow characteristics, walking patterns and behaviour are studied under varying land uses and at different types of facilities. All the study locations are in urban areas with high pedestrian activities. Videography was used to collect the data at these study locations. The data were later manually extracted and various analyses on pedestrian speed, flow and behaviour are performed. Pedestrian perception survey was also conducted to understand the perception of pedestrians towards various influencing factors affecting them. Pedestrian speeds are studied under the influence of various factors like facility type, pedestrian's own characteristics, involvement in an activity, group characteristics, temporal variations, land uses, bottlenecks and frictions, geographical location and IV vehicular traffic. Pedestrians in India are found walking at an average speed of 69.90 m/min. The type of facility is found to have significant influence on the pedestrian walking speeds. Pedestrians walk faster on sidewalks and carriageway and are slowest on precincts. Speeds of pedestrians in different age groups are found to vary significantly. Even though males walk faster than females, there was no significant difference in their walking speeds irrespective of type of facility. The usage of cell phones significantly reduced the walking speeds of pedestrians on all pedestrian facilities considered in this study. Pedestrians carrying baggage are found to walk significantly different from those without baggage on sidewalks, whereas, the difference is not significant in rest ofthe facilities. It is found that the pedestrian speeds vary significantly for pedestrian groups of different sizes. The composition of the group like when moving in afamily, walking with an infant and walking as pedestrian pairs also significantly influences the walking speeds. Pedestrians in different land uses walk significantly different with pedestrians in educational land use being faster and those in shopping land use being slower. Bottlenecks have an adverse effect on the pedestrian walking speeds. Speeds of exercise walkers are significantly faster than those of leisure walkers. Pedestrians in South India are walking significantly slower than those in North India. Anew method of estimating the influence of various factors on pedestrian speed is suggested in the present research so that the design pedestrian volume can be adjusted accordingly. The adjusted pedestrian volume will help in economizing and improving the efficiency of the pedestrian facilities. The flow characteristics of pedestrians are analyzed for different types of facilities and land uses. The basic speed - density relationship is found to be exponential for exclusive pedestrian facilities and linear for carriageway. Free flow speed of pedestrians is higher on sidewalks (94.60 m/min) and lower on precincts (80.38 m/min). Maximum flow rate (110. 91 ped/min or 6655 ped/hour) is found to occur on carriageway. It is observed that as the friction increases on the carriageway, pedestrians walk faster to get out of that situation as soon as possible. Pedestrian flow characteristics are found to vary for different land uses. Depending on the type of pedestrian facility, pedestrian flow characteristics vary within the same land use. In order to minimize the interactions with the vehicular traffic on carriageway, pedestrians are observed to walk quite close to each other leading to a higher density and lower space requirement on carriageway in mixed land use as compared to other land uses. Area module under maximum flow is found to vary between 0.32 and 0.58 m /ped for different land uses. It is found that the flow characteristics of Indian pedestrians are different from their counterparts in Western countries and hence the design guidelines recommended by HCM 2000 for design of pedestrian facility cannot be used. The walking pattern of pedestrians under the influences of various factors like bottleneck, encroachment, hawkers, vendors, presence of barrier and buffer, shops and surface irregularities revealed that pedestrians adopt different walking patterns based on the local conditions. Irrespective of the gender, 75% of the pedestrians avoided bottleneck by shifting on to the either side of the obstruction, if possible or by shifting on to the carriageway. Further, the presence or absence of the pedestrian facility alters the walking pattern of pedestrians on carriageway. As the traffic volume increased from 1457 veh/hour to 4875 veh/hour, percentage of pedestrians walking on the carriageway decreased from 65% to 6%. The effect of traffic volume is more on the pedestrians walking outside 4.2 m from the edge of the carriageway than those walking inside 1.8 m from the edge of the VI carriageway. Chances of pedestrians' interaction with the vehicular traffic have increased by 31% and 51% because of a parked cycle rickshaw and car respectively on the carriageway. The behaviour of pedestrians is found to varies with type of facility, personal characteristics like age and gender, involvement in an activity like carrying baggage and using cell phones while walking, purpose of walking like exercise and leisure, size and composition of the group, and land use. Pedestrian groups' behaviour is found to be different from at of individual pedestrians. The results of the pedestrian perception survey revealed that pedestrians perceive size of the facility, safety and convenience as the most important factors for design of a facility. Pedestrians waking on the carriageway consider parked vehicles, vehicular traffic on the carriageway, and width available on the carriageway as major factors for designing a new facility. Initially 22 parameters were identified for LOS model. Some of them were eliminated based on the Pearson correlation coefficient and partial correlation between the pedestrian LOS and the factors. The factors affecting the pedestrian LOS significantly are congestion (pedestrian flow), width of the facility, width occupied by encroachment, hawkers, and parked vehicle, number of lanes on the carriageway, cross flow and type of land use. The step wise regression analysis was employed to develop the pedestrian LOS model. Various combinations of the significant factors were tried and the best fit model for sidewalks, wide-sidewalks and precincts are developed. These equations are found to be statistically sound. Based on these equations, the design guidelines for different types of pedestrian facilities in varying land uses are recommended. The design guidelines for pedestrians in India are found to be different from those suggested in HCM 2000. Since the vii models developed in the present study incorporate the type of pedestrian facility and land use, the design guidelines for a particular type of facility in a specific land use can be obtained. This will aid in better design of pedestrian facilities, which is of utmost importance for a developing country like India. VIII