Microsoft Word - ETASR_V13_N4_pp11157-11165_needs_update Engineering, Technology & Applied Science Research Vol. 13, No. 4, 2023, 11157-11165 11157 www.etasr.com Srivastava & Kumar: The Impact of Road Side Friction on the Traffic Flow of Arterial Roads in Varanasi The Impact of Road Side Friction on the Traffic Flow of Arterial Roads in Varanasi Kartik Srivastava Department of Architecture and Planning, National Institute of Technology Patna, India kartikshri128@gmail.com (corresponding author) Ajay Kumar Department of Architecture and Planning, National Institute of Technology Patna, India arajay@nitp.ac.in Received: 1 April 2023 | Revised: 19 April 2023 | Accepted: 27 April 2023 Licensed under a CC-BY 4.0 license | Copyright (c) by the authors | DOI: https://doi.org/10.48084/etasr.5897 ABSTRACT Varanasi's prominence as a cultural and historical center means it receives visitors from all over the world. The city's tourism industry is a boon to the local economy. Better transportation infrastructure is crucial to attracting more tourists and increasing revenue. Varanasi is currently experiencing terrible difficulties due to various roadside frictional activities. Vehicle ownership grows in tandem with population growth. The increase in the number of vehicles on the road significantly impacts the reliability of the transportation network since land availability is fixed. The various roadside frictional activities usually found in the streets of Varanasi are on-street parking, pedestrian crossing, and Non-Motorized Vehicles (NMVs). There are not enough legal on-street parking and segregated lanes for NMVs/slow-moving vehicles or demarcation for pedestrian movements in the old city of Varanasi. Vehicles in Varanasi are traditionally parked on the street due to the narrow carriageways. Slow-moving vehicles are forced to move with fast-moving vehicles, and pedestrian crossings affect traffic flow. Integrated movement of slow- moving vehicles and rapid-moving vehicles affects traffic speed, pedestrian crossings impact the Level of Service (LOS), and on-street parking results in the reduction of the effective carriageway width and, hence, road capacity. This paper aims to identify the impact of pedestrian crossings, NMVs, and on-street parking on the traffic flow, speed, capacity, and LOS of urban arterial roads in Varanasi. To achieve this objective, two case studies were considered: the base section (with minimum side friction) and the friction section (with maximum side friction). The videography method was used for data collection. Nine hours of data were collected from 9:30 am to 6:30 pm. The video was played on the screen for data extraction. The speed model was developed by using fundamental diagram methods. Speed-density curve was drawn using Greenshield’s model. The speed-flow curve was derived from the speed-density curve to estimate the capacity at the base and friction sections. Reduction in capacity was determined by comparing friction section capacity with base section capacity. V/C ratio of a particular road compared with the V/C value provided by IRC 106 to predict LOS. A correlation model was developed between the percentage reduction in capacity and road width. Increase in parking, the proportion of NMVs, and pedestrian crossing frequency reduce the traffic flow. It was observed that a 34.10% capacity loss occurs at a 29.31% reduction in effective width, and a 40.54% loss occurs at a 40.98% reduction in effective width. Roads with frictional activities affected its level of service up to LOS-E (for the Bhelupur section). Keywords-side friction; speed; capacity; level of service; non-motorized vehicles; on-street parking; pedestrian crossing; traffic flow; linear regression; correlation I. INTRODUCTION Side frictional activities take place on or along the road and disrupt traffic flow. Varanasi, the cultural capital of India, attracts visitors from all over the globe. On-street parking, non- motorized transportation, and pedestrian crossing are all activities that add friction to the streets of Varanasi. These activities affect traffic flow characteristics in several ways: they result in road capacity reduction, impact speed, and affect the Level of Service (LOS). A vehicle requires parking space at the start and end of its journey. The significant growth in vehicular traffic has increased the need for parking in urban centers. Scarce parking can interrupt traffic flow and produce congestion. In Varanasi, on-street parking is a commonality, since there simply aren't enough legal on-street parking places, so people use to park their vehicles on the street, hence inducing a reduction in effective carriageway width. Also, there is no segregated lane for non-motorized transport, which forces slow- and fast-moving vehicles to flow in the same lane, which causes a reduction in traffic flow speed and LOS. Pedestrian crossing is another common problem in Varanasi. Engineering, Technology & Applied Science Research Vol. 13, No. 4, 2023, 11157-11165 11158 www.etasr.com Srivastava & Kumar: The Impact of Road Side Friction on the Traffic Flow of Arterial Roads in Varanasi II. LITERATURE REVIEW In [1], various side friction elements were considered, such as on-street parking, city buses stopping on the roadway, exit/entry vehicles, and vehicles about to turn. The study proposed a novel formula to update the Indonesian highway capacity manual to estimate the speed and capacity of urban highways with significant side friction. In [2, 3], several forms of roadside frictional parameters were discussed along with their impact on urban road capacity, highlighting the relationship between side friction parameters, road capacity, and LOS. These studies showed that capacity reduction in urban streets depends on roadside activity and varies with side friction. Loss of capacity on urban highways is caused by various factors, including roadside activity such as on-street parking, pedestrian movement, the presence of street vendors, the lack of lateral clearance and lane discipline, and diverse traffic, which is particularly prevalent in developing countries like India. In [4], a microscopic simulation model was developed to investigate the impact of bus stops on mixed traffic movement, focusing on reducing traffic flow rates. This study evaluated the impact of bus stops, which act as side friction on urban traffic. Based on the collected data, several areas for further work were recommended, as very few studies examine mixed traffic situations [5, 6]. In [7], three different side friction factors were used in a survey (on-street parking, bus bay stops, and curbside bus stops) and their impact on traffic quality performance was examined. This study showed an average speed drop due to side friction, a reduction in stream speed of 49-57% due to bus stops and bays, and a loss of 45-67% in speed due to on-street parking. This study used dynamic and static PCU (Passenger Car Unit) values. As a result of using these two different types of PCU values resulted in a 10-53% capacity reduction in bus bays and bus stops and about 28-63% due to on-street parking. Authors in [8-10] investigated urban arterial roads, finding that parked and stopped vehicles act as frictional elements that impact traffic flow, LOS, speed, and delay. On-street parking has been highlighted as a substantial source of side friction in urban arterial roads, and there is a relationship between the presence of on-street parking and the reduction in stream speed. According to [11-13], the impact of on-street parking as a side friction element on urban arterial roads results in a slowdown in traffic flow ranging from 15 to 44% or 5.1 to 21 km/h. In [13], parking density was considered a side friction element. The survey results showed that parking density on urban arterial roads harms average traffic flow, with a decrease of around 13 km/h for every 100 vehicles/km increase in parking density. In [15], various types of side frictional activities were examined, using small simulation models to evaluate how bus stops impact mixed traffic movement and their relation to the reduction in travel speed. In [2, 8], parked and stopped automobiles were studied, along with various other factors. These studies showed that parked and stopped vehicles are critical side frictional elements, as they harm the regular traffic flow, speed, capacity, and LOS more than other side frictional factors, such as pedestrian movements. In [9, 14], T-junctions and bus stops were characterized as static features contributing to traffic congestion. On the other hand, static factors can be addressed by upgrading the geometric aspects of roadways to increase LOS. Roadside friction elements found in urban roads in Asian countries were pedestrians, stopped vehicles, parking, and access to roadside premises. Based on the weights assigned to each side friction element, side friction classes were developed for urban and interurban roads. In [16], motorcycle taxis were found to add another degree of friction when parking or loading and unloading passengers. In [17], on-street parking produced a decrease in stream speed, ranging from 15 to 44% or 5.1 to 21 km/h. In [18], speed–flow diagrams at various flow levels were plotted to analyze the collective effect of all these factors. The capacity of urban motorways and four-lane two ways were found as 4600 PCU/ and 5700 PCU/h, which further matched with the capacity values recognized in the South Korean manual and US HCM. The study concluded that Indonesian roads function poorer than those in the USA and South Korea. Drivers' speed choices are expressively affected by crossing pedestrians, engaged curbside bus stops and confrontation with vehicles in the opposing direction [19]. The effect of crossing pedestrians on stream speed was analyzed using the network- based speed–flow relationships by [20]. Authors in [21] established a microscopic simulation model in VISSIM to simulate pedestrian flow properties by coding pedestrians as a vehicle that follows the pedestrian flow characteristics despite many issues related to pedestrian behavior, pedestrian route choice decisions, and multidirectional flow. Authors in [22] demonstrated the effect of operating speed on the capacity of a midblock section of urban road using field data collected at 12 midblock sections in India. Authors in [23] evaluated the effect of the pedestrian refuge island on the speed of vehicles and, accordingly, the effects on the probability variations of fatal accidents. Authors in [24] used the simulation approach to study pedestrian crossing behaviors. They concluded that pedestrians show more tolerance in waiting time in developing countries as compared to developed countries. Due to the high percentage of heavy vehicles, this study used dynamic and static PCU values. As a result, a 10-53% capacity reduction was observed in bus bays and bus stops and about 28-63% was attributed to on-street parking. Authors in [25] included NMVs (Non-Motorized Vehicles) as a source of side friction. He created a speed model for this purpose, which is based on the ratio of NMVs. A 10% rise in the NMV ratio, according to the model, would result in a 0.8 mile per hour (1.29 km/h) reduction in average traffic flow speed. The inclusion of NMVs in the traffic volume classification was suggested because cycle rickshaws generate greater side friction than bicycles. Authors in [26, 27] identified the degree of correlation between NMVs and traffic performance. The side frictional element (NMVs) is further divided into various subcategories, such as the proportion of NMVs, the proportion of bicycles, the proportion of cycle rickshaws, the proportion of handcarts, and the percentage of NMVs that travel beyond 20% of the roadway width from the edge. The correlation values of N-3 and N-4 were the highest among all the subcategories. The literature review reveals that most researchers tried to find the impact of various side frictional activities on traffic characteristics individually. Only some studies have been carried out on more than one and maximum two side friction parameters combined. To the best of our knowledge, no study Engineering, Technology & Applied Science Research Vol. 13, No. 4, 2023, 11157-11165 11159 www.etasr.com Srivastava & Kumar: The Impact of Road Side Friction on the Traffic Flow of Arterial Roads in Varanasi related to the combined effect of on-street parking, NMV, and pedestrian crossing on traffic flow, speed, capacity, and LOS has been attempted for mixed traffic conditions. The present research attempts to identify the impact of these 3 side frictional parameters on various traffic characteristics like traffic flow, speed, capacity, and LOS. III. RESEARCH OBJECTIVES The present work aims to study the impact of on-street parking, pedestrian crossing, and NMVs on traffic flow, traffic speed, road capacity, and LOS on the urban arterial roads of Varanasi. IV. METHODOLOGY Study stretch selection: To achieve the objective of this research, firstly, two types of study stretches were chosen: Base Section (BS), with minimum side friction and Friction Section (FS), with maximum side friction. These sections are away from the intersection to avoid its effect. Seven BSs and seven FSs were selected based on the similarity of road characteristics. Planning phase: in this stage, two survey methods were prepared. The first one is a manual survey to collect geometric details of the road. The survey format (traffic volume count, parking, speed, and pedestrian movements) were prepared to observe data manually. The second is the videographic method. A 60 m stretch was marked to be captured in the video. The camera was mounted at a certain height and the time of survey was recorded. Data collection: Data were collected on both stretches for 9 hours per day, from 9:30 am to 6:30 pm. The data were collected from 14 mid-block sections (7 BSs and 7 FSs) of a 2- lane urban arterial road. Data regarding classified traffic volume count, peak hour volume, peak hour duration, number of on-street parking, parking duration, speed of traffic flow, traffic density, and to examine the LOS were extracted from the recorded videos. Data extraction: The recorded videos were played on screen, Individual vehicles were counted at an interval of 15 min. These data were used to calculate the classified traffic volume count, proportion of non-motorized vehicles, peak hour volume, peak hour factors, peak hour duration and heterogeneous traffic flow. To understand the on-street parking demand during peak hours, the parking vehicles were counted and the parking duration was noted. The pedestrian crossings during peak hours were counted. The video data were used for speed measurements. The entry and exit times of every vehicle over a 60 m trap length were observed. Speed was calculated for the individual vehicles by dividing the 60 m trap length with the entry-exit time difference of a vehicle. Data analysis: By using peak hour factor, traffic flows at BSs and FSs were calculated. The FS traffic flow rate was compared with BSs' to determine the reduction in traffic flow rate. The video data were used for speed measurement. Individual vehicles' entry and exit times over a 60 m trap length were observed. Speed was calculated for individual vehicles by dividing the entry-exit time difference of a vehicle by 60 m trap length. The obtained space mean speed of the vehicles was used to calculate the traffic speed. The data set of speed and flow for all base and friction sections were pooled together and a speed prediction model was developed. By using the speed- flow curve, the capacity for each parking number and NMV proportion was determined. All the observed capacities at the base and friction sections were compared, and the reduced capacity due to on-street parking and NMVs is calculated. After this, the parking numbers were correlated with the available effective width and a relationship between capacity reduction and available effective width was developed. Finally, the volume/capacity (V/C) ratios were calculated for the FS. The value obtained from the V/C ratio was compared with the V/C ratio provided by the Indian Road Congress (IRC) to measure the LOS. V. DATA COLLECTION This study focuses on the impact of on-street parking on road capacity, speed, and LOS, the result of non-motorized transport on traffic flow speed and LOS, and the effect of pedestrian crossings on rate and LOS. To achieve the objective, seven midblock 2-lane arterial sections with minimum side frictions were selected. Indiranagar, Bisesarganj, Sonia, Salarpur, Lahurabeer, DLW campus and BHU (A1, B1, C1, D1, E1, F1 and G1), were named as the BSs. Similarly, seven midblock 2-lane arterial sections, with side frictions, were selected with different levels of parking maneuvers, NMV, and pedestrian crossing. Chitaipur, Godowlia, Benia, Sarnath, Sigra, DLW and Bhelupur (A2, B2, C2, D2, E2, F2 and G2), respectively, were named as the FSs. Both manual and videographic methods were used for data collection. The data were collected for 9 hours, from 9:30 am to 6:30 pm, on both types of stretches (BS and FS) on weekdays only. A 60 m stretch was chosen for data collection. The global coordinates of each roads are: Indira nagar (25°16'25.77" N, 82°58'7.34" E), Bisesarganj (25°19'26.89" N, 82°59'30.05" E, Sonia (25°21'14.49" N, 83°1'25.04" E), Salarpur (25°21'12.07" N, 83°2'19.76" E), Lahurabeer (25°19'8.44" N, 83°0'5.23" E), DLW campus (25°17'11.05" N, 82°58'9.21" E), BHU(25°16'3.79" N, 82°59'28.53" E), Chitaipur (25°16'18.72" N, 82°58'3.47" E), Godowlia (25°18'36.27" N, 83°0'34.74" E), Benia (25°19'3.28" N, 83°0'15.89" E), Sarnath (25°22'34.2" N, 83°1'21.76" E), Sigra (25°18'39.89" N, 82°59'11.19" E), DLW (25°17'11.05" N, 82°58'9.21" E), and Bhelupur (25°18'14.29" N, 82°59'24.03" E). From the inventory survey, geometric details of the road section, carriageway width, right of way, on- street parking, road surface quality, street vendors etc., were obtained. The classified traffic volume count survey was conducted manually (survey format) and by videographic survey. The data were collected at 15 min intervals. From the traffic volume count survey, traffic composition, peak hour traffic, peak hour volume, directional split, peak hour factor, and proportion of motorized and non-motorized vehicles were all measured to learn more about the traffic characteristics of each place. For counting flow, a video is played, and each vehicle category is counted per min. The video data were used for speed measurements for a 60 m trap length entry and the exit time of vehicles was noted. In order to calculate the speed of individual vehicles, the 60 m trap length was divided by the interval exit time – entry time. The same procedure was applied Engineering, Technology & Applied Science Research Vol. 13, No. 4, 2023, 11157-11165 11160 www.etasr.com Srivastava & Kumar: The Impact of Road Side Friction on the Traffic Flow of Arterial Roads in Varanasi for both BS and FS. From the parking survey, the number of parking maneuvers was calculated for the peak hours of the respective sections, the parking duration, the types of parked vehicles, the peak hour parking demand, the occupancy level, and the proportion of the road occupied by on-street parking were noted. From the pedestrian survey, the number of pedestrian crossings during peak hours was observed in 15 min intervals. Tables I and II represent the vehicular composition of the BSs and FSs chosen in this study. The percentages of motorized and NMVs are calculated from these data and are used in further analysis. As the traffic is heterogeneous, in order to convert the stream into homogenous traffic, the data were used with PCU to get homogenous traffic conditions. This process simplifies the analysis phase. Tables III and IV represent the traffic volume count at the BSs and FSs. The collected data were used to calculate peak hour (from 9:30 am to 11:30 am) volume, as shown in Table XI. A parking survey was conducted at both sections by videographic survey from 9:30 am to 6:30 pm. It can be observed in Table V that all roads have on-street parking issues. The parking location on each road shows all roads having an unauthorized type of parking. They determine the on-street parking demand during peak hours and identify its impact on speed, capacity, and LOS. The parking maneuvers during peak hours in various vehicle categories are shown in Table VI. VI. RESULTS AND DISCUSSION A. Impact of On-street parking, NMV, and Pedestrian crossing on Traffic Flow The peak time of the BS (the hour of the day with the highest traffic volume for that section) can be seen in Tables III-IV. The peak hour volumes of the BSs and the FSs are shown in TableVII. TABLE I. VEHICULAR COMPOSITION AT THE BASE SECTIONS BS Νame Vehicular composition (in % age) Car 2W 3W Bus/truck SGV LGV Cycle Others A1 Indira nagar 23.97 36.03 22.37 3.26 7.63 4.74 0.33 1.67 B1 Bisesarganj 18.69 39.37 19.32 5.68 6.21 7.23 1.47 2.03 C1 Sonia 18.36 41.12 20.87 4.96 5.26 6.66 1.33 1.98 D1 Salarpur 24.32 39.67 19.46 5.32 4.21 4.66 1.20 1.16 E1 Lahurabeer 22.96 40.48 17.56 5.03 4.52 5.72 1.60 2.09 F1 DLWcampus 28.66 42.39 15.23 2.01 5.62 0.92 3.37 1.83 G1 BHU 22.84 42.98 18.76 3.20 3.96 1.82 5.23 1.21 * SGV = small goods vehicle, LGV = large goods vehicle TABLE II. VEHICULAR COMPOSITION AT THE FRICTION SECTIOΝS FS Name Vehicular composition (in % age) Car 2W 3W Bus/truck SGV LGV Cycle Others A2 Chitaipur 25.34 38.66 20.09 4.91 3.26 4.74 0.87 2.13 B2 Godowlia 13.86 42.24 21.14 2.00 8.76 7.29 1.83 2.88 C2 Benia 17.28 43.13 22.18 2.90 7.87 2.10 1.60 2.40 D2 Sarnath 26.21 38.97 19.23 4.77 3.97 5.03 0.80 1.20 E2 Sigra 23.07 37.93 20.91 6.09 3.26 5.74 1.76 1.24 F2 DLW 29.87 41.13 13.66 2.34 7.98 1.03 2.02 1.97 G2 Bhelupur 21.07 44.93 19.02 2.79 4.21 1.90 4.98 1.10 TABLE III. TRAFFIC VOLUME COUNTS AT THE BASE SECTION BS 9:30-9:45 9:46-10:00 10:01-10:15 10:16-10:30 10:31-10:45 10:46-11:00 11:01-11:15 11:16-11:30 A1 277 232 201 211 159 248 152 176 B1 348 346 350 337 232 221 388 279 C1 233 247 242 239 198 202 187 203 D1 339 347 350 344 336 297 252 289 E1 296 289 250 232 249 286 192 132 F1 342 375 326 266 298 287 232 198 G1 226 268 275 266 295 286 176 145 TABLE IV. TRAFFIC VOLUME COUNT AT THE FRICTION SECTION FS 9:30-9:45 9:46-10:00 10:01-10:15 10:16-10:30 10:31-10:45 10:46-11:00 11:01-11:15 11:16-11:30 A2 154 146 186 183 198 181 163 159 B2 282 276 206 216 211 211 142 144 C2 186 158 142 153 182 182 127 113 D2 289 248 252 244 179 191 188 136 E2 141 137 143 141 139 127 132 138 F2 214 273 268 152 189 197 157 196 G2 108 113 137 109 111 88 98 108 Engineering, Technology & Applied Science Research Vol. 13, No. 4, 2023, 11157-11165 11161 www.etasr.com Srivastava & Kumar: The Impact of Road Side Friction on the Traffic Flow of Arterial Roads in Varanasi TABLE V. ON-STREET PARKING LOCATIONS BS Place Type FS Place Type LHS RHS LHS RHS A1 * Un-Authorized A2 * Un-Authorized B1 * Un-Authorized B2 * Un-Authorized C1 * Un-Authorized C2 * Un-Authorized D1 * Un-Authorized D2 * Un-Authorized E1 Un-Authorized E2 * Un-Authorized F1 * * Un-Authorized F2 * Un-Authorized G1 * Un-Authorized G2 * Un-Authorized * LHS = left hand side, RHS = right hand side TABLE VI. PARKING MANOEUVRES BS Parking manoeuvres (Peak hours) Car 2w Auto LCV Truck Other Total A1 1 2 2 - - 2 7 B1 - 4 1 - - 5 C1 1 1 - - - 1 3 D1 1 3 1 - - 2 7 E1 - 1 - - - - 1 F1 - 1 1 - - - 2 G1 - 1 - - - 1 2 A2 13 36 17 3 1 6 76 B2 18 27 17 7 - 4 73 C2 8 24 32 12 - 8 84 D2 12 32 16 9 2 6 80 E2 16 36 18 3 2 9 84 F2 18 32 21 7 - 11 89 G2 16 28 15 5 - 9 73 TABLE VII. PEAK HOUR VOLUME BS Peak hour volume FS Peak hour volume A1 921 A2 748 B1 1381 B2 980 C1 961 C2 639 D1 1380 D2 1033 E1 1067 E2 562 F1 1309 F2 907 G1 1122 G2 470 B. Peak Hour Factor (PHF) PHF is the hourly volume during the analysis hour divided by the 4 × peak 15-minute flow rate within the analysis hour. The PHF is also considered a measure to estimate traffic demand fluctuations and traffic flow rate within the peak hour. PHF is used to calculate the base and friction section traffic flow rate during peak hours. As shown in Table IX, the traffic flow rate can be calculated by peak hour volume with the PHF. Table X shows that due to side frictional activities, the reduction percentage in traffic flow ranged from 36.67% to 66.13%. Bhelupur stretch is majorly impacted by side frictional activities as a reduction in traffic flow is 66.13%, while the Chitaipur stretch is the least affected with a reduction in traffic flow of 36.67%. As shown in Figure 1, due to various side frictional activities at the FS with respect to the BS (free from any side friction), the reduction in traffic flow can be observed due to on-street parking, NMVs and pedestrian crossing are directly proportional to the reduction percentage in traffic flow. The R 2 value of 0. 896 for the created model indicates high dependability. TABLE VIII. TRAFFIC FLOW RATE BS PHF at the BS Traffic flow rate at the BS (Veh/hr) A1 0.83 1110 B1 0.98 1410 C1 0.97 991 D1 0.98 1409 E1 0.90 1186 F1 0.87 1505 G1 0.95 1181 FS PHF at the FS Traffic flow rate at the FS (Veh/hr) A2 0.94 703 B2 0.86 843 C2 0.85 543 D2 0.89 919 E2 0.98 551 F2 0.83 753 G2 0.85 400 TABLE IX. REDUCTION PERCENTAGE IN TRAFFIC FLOW DUE TO FRICTIONAL ACTIVITIES Reduction in traffic flow (Veh/h)r Traffic flow reduction percentage (%) 407 36.67 567 40.21 448 45.20 490 34.78 635 53.54 750 49.83 781 66.13 Fig. 1. Flow-density relationship at FS C. Speed-Density Relationship Three fundamental parameters, traffic flow, speed and density, were used to develop the relationship between traffic speed and density of the 14 sections. Space mean speed data extracted from videography were used to predict traffic speed. The speed of traffic flow in each section was determined using the 85 th percentile cumulative method as in (1). Traffic volume data were converted into passenger car units to develop a speed-flow relationship. Each vehicle type was found in mixed traffic conditions, so it was more challenging to predict the density and capacity of the road. To analyze the mixed scenario space, the mean speed used in this research. Space Mean Speed (SMS) of vehicles was noted in the presence of on-street parking and NMVs. The SMS for the Indiranagar section (A1) is shown in Table XI. The values of the speed of traffic flow at the BS and the FS in the presence of on-street parking and NMVs are shown in Table XII. Table XIII and Figure 2 show the reduction in speed in the FSs. The R 2 value of 0.865 for the Engineering, Technology & Applied Science Research Vol. 13, No. 4, 2023, 11157-11165 11162 www.etasr.com Srivastava & Kumar: The Impact of Road Side Friction on the Traffic Flow of Arterial Roads in Varanasi created model indicates high dependability. These speed data were further used in the prediction of road capacity. TABLE X. TRAFFIC SPEED AT BASE AND FRICTION SECTION BS Traffic speed at BS FS Traffic speed at FS A1 29.081 km/hr A2 13.392 km/hr B1 32.173 km/hr B2 14.637 km/hr C1 32.112 km/hr C2 13.338 km/hr D1 30.694 km/hr D2 14.194 km/hr E1 29.963 km/hr E2 16.196 km/hr F1 27.252 km/hr F2 16.826 km/hr G1 29.246 km/hr G2 14.324 km/hr TABLE XI. TRAFFIC FLOW SPEED Parking manoeuvres at BS Pedestrian crossing at BS NMV at BS Parking manoeuvres at FS NMV at FS Traffic speed (Km/hr) BS FS 7 11 9 76 23 29.081 13.392 5 8 4 73 46 32.173 14.637 3 13 12 84 26 32.112 13.338 7 9 7 80 21 30.693 14.194 1 9 11 84 17 29.963 16.196 2 2 13 89 36 27.252 16.826 2 7 19 73 47 29.246 14.324 1 n SMS Vi   (1) where SMS is the space mean speed, n is the number of vehicles, and Vi is the speed of an individual vehicle. TABLE XII. REDUCTION IN SPEED AT THE FS DUE TO ON- STREET PARKING AND NMVs FS Speed reduction (kmph) Speed reduction percentage A2 15.689 53.94 B2 17.536 54.50 C2 18.774 58.46 D2 16.499 53.75 E2 13.767 45.94 F2 10.426 38.25 G2 14.922 51.02 Fig. 2. Speed reduction percentage in traffic speed at the FSs. D. Impact of On-street Parking and NMVs on Road Capacity Two types of data were used to determine the road section capacity at both sections: SMS and classified volume data. Data were extracted by the videography method at intervals of 15 min. A 60 m trap length was chosen for all 14 sections to collect data regarding speed and classified traffic volume count. A video camera was fixed 5.2 m above the ground, and 9 hours of video were recorded. The fundamental diagram method was used to establish the speed-flow relationship. Volume data (veh/h) were recorded in the video, which need to be converted into PCU as it was mixed traffic. Highway capacity manual categorized vehicle types should be considered as motorized and NMVs (cycles, cycle rikshaws, and animal-drawn vehicles). The PCU factors for individual vehicle categories given by IRC 106 (for urban streets) were used. Using PCU factors, each 15 min traffic count (veh/h) was converted into PCU/h and was multiplied by 4 to get the hourly volume. The flow (PCU/hr) at the BSs and FSs is shown in Table XIV. The density of each section was calculated by: � = �� (2) where � = �/�, Q is the traffic flow (veh/hr), K is the traffic density (veh/km), and V is thee traffic speed (km/hr). By using the Greenshield’s Model (3), speed and density graphs were designed to estimate the section capacity. Figure 3 shows the speed-density curve for the BSs. TABLE XIII. FLOW (PCU/H) BS/FS Vehicular composition in percentage Car 2W 3W Bus/truck SGV LGV Cycle Others Total A1 66 75 124 31.5 42 28.6 0.4 10 377.5 B1 65 103.5 136 38.5 44 55 2 14 458 C1 45 76.5 104 42 26 35.2 1.2 10 339.9 D1 85 104.25 136 49 46 33 0.4 8 461.65 E1 68 90 104 35 52 50.6 2.8 12 414.4 F1 96 119.25 82 45.5 38 46.2 2.4 16 445.35 G1 67 94.5 110 59.5 42 41.8 1.6 9 425.4 A2 48 57 78 10.5 26 15.4 0.4 2 237.3 B2 43 73.5 102 17.5 30 28.6 0.8 6 301.4 C2 26 57 78 24.5 18 24.2 0.4 4 232.1 D2 63 76.5 102 14 26 19.8 0.8 2 304.1 E2 49 64.5 66 7 24 26 2.4 6 244.9 F2 76 99 58 14 18 15.4 1.6 8 290 G2 52 72 70 21 16 17.6 0.4 4 253 Engineering, Technology & Applied Science Research Vol. 13, No. 4, 2023, 11157-11165 11163 www.etasr.com Srivastava & Kumar: The Impact of Road Side Friction on the Traffic Flow of Arterial Roads in Varanasi TABLE XIV. FLOW, SPEED, AND DENSITY BS Flow (PCU /h) Stream speed (km/h) Density (PCU/km) FS Flow (PCU/h) Stream speed (km/h) Density (PCU/km) A1 1510 29.081 51.92 A2 950 13.392 70.93 B1 1832 32.173 56.94 B2 1206 14.637 82.39 C1 1360 32.112 43.71 C2 929 13.338 69.65 D1 1847 30.693 60.17 D2 1217 14.194 85.74 E1 1658 29.963 55.33 E2 980 16.196 65.50 F1 1782 27.252 65.38 F2 1160 16.826 68.94 G1 1702 29.246 58.19 G2 1012 14.324 70.65 Greenshield's straight line can be seen in the graph, best fitted on the basis of the R 2 value. Similarly, Figure 4 shows the speed-density graph for the FSs, while also the Greenshield's straight line can be observed, which is best fitted on the basis of the R 2 value. � = � – ∗ � (3) Fig. 3. Speed-density relationship at the BS. Fig. 4. Speed-density relationship at the FS. Table XV, shows the traffic flow, speed, and density at the BSs and FSs. To determine the capacity, the developed model (speed-density model) was used to generate the speed-flow curve shown in Figure 5. The speed-flow curve will give the capacity of the respective sections. The generated capacity is shown in Table XVI. The reduction percentage in the capacity due to on-street parking and NMV on the FS are shown in Table XVII. The LOS of each section was calculated by dividing the road volume with the capacity. The volume/capacity ratio was compared with the values provided by IRC 106. Table XVIII represents the impact of on-street parking, NMV, and pedestrian crossing on LOS at the FS. According to the IRC guidelines for urban roads, a road will lie in LOS A if its V/C value lies in 0-0.60 (free flow), similarly for LOS B, it should be in 0.61-0.70 (reasonable free flow), for LOS C 0.71-0.80 (near free flow), for LOS D 0.81- 0.90 (medium flow), for LOS E 0.91-1.00 (capacity flow), and LOS F values should be greater than 1 (congested flow). It can be observed from Table XVIII, that frictional activity affects the service level of Sarnath road up to LOS-B, while the level of service of Bhelupur road is impacted up to LOS-E. Fig. 5. Correlation between the reduction percentage, effective width, and capacity. TABLE XV. CAPACITY REDUCTION PERCENTAGE DUE TO NMVs Capacity at BS (PCU/hr/2lane) Capacity at FS (PCU/hr/2lane) Reduction in capacity (PCU/hr/2lane) Reduction percentage (%) 1510 950 560 37.08 1832 1206 626 34.17 1360 929 431 31.69 1847 1217 630 34.10 1658 980 678 40.86 1782 1160 622 34.90 1702 1012 690 40.54 TABLE XVI. CAPACITY REDUCTION PERCENTAGE DUE TO ON-STREET PARKING Parking manoeuvres/hr NMT Reduction percentage (%) 76 23 37.08 73 46 34.17 84 26 31.69 80 21 34.10 84 17 40.86 89 36 34.90 73 47 40.54 TABLE XVII. IMPACT ON LOS AT FS FS V/C at FS LOS as per IRC Remarks A2 0.787 LOS C Near free flow B2 0.812 LOS D Medium flow C2 0.687 LOS D Medium flow D2 0.848 LOS B Reasonable free flow E2 0.873 LOS D Medium flow F2 0.781 LOS C Mear free flow G2 0.902 LOS E Capacity flow Engineering, Technology & Applied Science Research Vol. 13, No. 4, 2023, 11157-11165 11164 www.etasr.com Srivastava & Kumar: The Impact of Road Side Friction on the Traffic Flow of Arterial Roads in Varanasi E. Correlation of Capacity Reduction with the Effective Width of the Carriageway On-street parking along the roads led to a loss of carriageway width, resulting in loss of road capacity. A correlation model was developed between the effective carriageway width and the capacity of the FS. The available effective width is the width of the carriageway available at any instance for traffic movement. On-street parking reduces the available effective width, and the capacity of the friction section decreases, as shown in Table XIX. TABLE XVIII. RELATIONSHIP BETWEEN PARKING MANEUVERS PER HOUR, CAPACITY REDUCTION, AND EFFECTIVE WIDTH FS Parking manoeuvres per hour Capacity reduction (pcu/hr) Capacity Reduction percentage Flow at BS (veh./ 5 min.) A2 76 560 37.08% 126 B2 73 626 34.17% 153 C2 84 431 31.69% 113 D2 80 630 34.10% 154 E2 84 678 40.86% 138 F2 89 622 34.90% 149 G2 73 690 40.54% 142 For data extraction, the BS (ex: A1) with minimum (as possible) and the FS (ex: A2) with the maximum number of on- street parking maneuver issues were selected. The flow ratio (veh / 5 min) at these two sections for each interval is determined with the corresponding number of parking maneuvers. The ratio of flow at section A1 to section A2 directly gives the effective width factor for each 5 min interval, as shown in (4). These effective width factors for various sections at different parking maneuvers were multiplied with the actual width of the road to get the available effective width, as shown in (5). The reduction in width can be calculated by deducting the available effective width for various parking maneuvers from the actual width of the road. 2 1 F(A ) EWF= F(A ) (4) AEW=EWF×AW (5) where EWF is the effective width factor, F(A1) is the traffic flow at A1, F(A2) the traffic flow at A2, AEW is the available effective width, and AW the actual width of the road. Fig. 6. Correlation between the reduction percentage, effective width, and capacity. As shown in Figure 6, a linear relationship exists between width decreases and capacity reductions, with an R 2 value of 0.853. It is determined that a 34.10% capacity loss occurs at a 29.31% reduction in effective width, and a 40.54% loss occurs at a 40.98% reduction in effective width. Reduced effective width and capacity are related, as demonstrated in (6). Percentage reduction in width = 0.709 × percentage reduction in capacity + 0.109 (6) with R 2 = 0.853. VII. CONCLUSION Roadside frictional activities are prevalent in developing countries like India. This study focused on the urban arterial roads in Varanasi. It is observed that the presence of side frictional activities such as on-street parking, NMVs, and pedestrian crossing reduces the traffic flow, speed, and road capacity and impacts LOS. In this study, two types of roads were considered: base section (with minimum side friction) and friction section (with maximum side friction). Data were collected from 7 base and 7 friction sections to analyze the effects of the considered fractional activities on the friction section. A correlation model was developed at the friction section and was compared with the base section to understand the impact of on-street parking, NMV, and pedestrian crossing on traffic flow. The speed-density model was developed using three fundamental parameters, which were further used to develop the speed-flow model. The speed-flow curve is plotted using the model's speed at various flow levels. The speed-flow curve is used to estimate the capacities of respective roads. Capacities at each parking maneuver and the proportion of NMV were determined and compared with the base section capacity. The capacity obtained for different roads was used to calculate the V/C ratio to understand the quality of the respective roads' service levels. The values obtained from the V/C ratio were compared with those suggested by IRC 106 guidelines. The main conclusions of this study are:  Traffic flow of two-lane divided friction urban arterial midblock sections decreases with the increase in the influence of on-street parking, NMV, and pedestrian crossing, compared to the base section. The R 2 value obtained from the developed model indicates high dependability.  The reduction percentage in traffic flow ranged from 34.78% to 66.13%. Bhelupur stretch is majorly impacted by side frictional activity as the reduction in traffic flow is 66.13%, while the Sarnath stretch is least affected with a reduction in traffic flow of 34.78%.  The R 2 value obtained from flow-density relationship at BS is 0.576, at FS it is 0.896, showing high dependability between traffic flow rate and side frictional parameters.  The reduction percentage in speed at friction section ranged from 38.25% to 58.46%. From the developed model between % age reduction in speed and parking maneuvers per hour, the value of R 2 (0.865) indicates high dependability. Engineering, Technology & Applied Science Research Vol. 13, No. 4, 2023, 11157-11165 11165 www.etasr.com Srivastava & Kumar: The Impact of Road Side Friction on the Traffic Flow of Arterial Roads in Varanasi  A correlation is obtained between the available effective width and percentage reduction in capacity.  It is observed that a 34.10% capacity loss occurs at a 29.31% reduction in effective width, and a 40.54% loss occurs at a 40.98% reduction in effective width.  The capacity of Indira nagar road (A1) is 1510 PCU/hr/2lane, while the capacity of Chitaipur Road (A2) is reduced to 950 PCU/hr/2lane due to on-street parking and NMV.  Similarly, the capacity of Lahurabeer road (E1) is 1658 PCU/hr/2lane, while the capacity of Sigra road (E2) is reduced to 980 PCU/hr/2lane due to the influence of frictional activities. REFERENCES [1] A. Munawar, "Speed and Capacity for Urban Roads, Indonesian Experience," Procedia - Social and Behavioral Sciences, vol. 16, pp. 382–387, Jan. 2011, https://doi.org/10.1016/j.sbspro.2011.04.459. [2] Y. Xiaobao, H. Mei, and G. Ziyou, "Car Delay Model near Bus Stops with Mixed Traffic Flow," Journal of Applied Mathematics, vol. 2013, May 2013, Art. no. e437637, https://doi.org/10.1155/2013/437637. [3] M. A. Lim, E. L. Hallare, and J. G. Briones, "Modeling the impact of on- street parking on vehicular traffic," in 5st Atrans Symposium Student Chapter Session, Bangkok, Thailand, Aug. 2012, pp. 47–57. [4] W. E. Marshall, N. W. Garrick, and G. Hansen, "Reassessing On-Street Parking," Transportation Research Record, vol. 2046, no. 1, pp. 45–52, Jan. 2008, https://doi.org/10.3141/2046-06. [5] S. Biswas, S. Chandra, and I. Ghosh, "Effects of On-Street Parking in Urban Context: A Critical Review," Transportation in Developing Economies, vol. 3, no. 1, Apr. 2017, Art. no. 10, https://doi.org/10.1007/s40890-017-0040-2. [6] S. Chand and S. Chandra, "Impact of bus stop on urban traffic characteristics–a review of recent findings," Journal of Society for Transportation and Traffic Studies, vol. 5, no. 2, pp. 57–72, 2014. [7] M. Patkar and A. Dhamaniya, "Effect of On-street parking on Effective Carriageway Width and Capacity of Urban Arterial Roads in India," European Transport, no. 73, 2019, Art. no. 1. [8] R. Khatawkar and C. Kumar N T, "Impact of Road Side Bus Stops on Capacity of Roadway," International Journal of Avanced Technology in Engineering and Science, vol. 3, no. 7, pp. 147–157, 2015. [9] H. Guo, Z. Gao, X. Yang, X. Zhao, and W. Wang, "Modeling Travel Time under the Influence of On-Street Parking," Journal of Transportation Engineering, vol. 138, no. 2, pp. 229–235, Feb. 2012, https://doi.org/10.1061/(ASCE)TE.1943-5436.0000319. [10] R. Z. Koshy and V. T. Arasan, "Influence of Bus Stops on Flow Characteristics of Mixed Traffic," Journal of Transportation Engineering, vol. 131, no. 8, pp. 640–643, Aug. 2005, https://doi.org/10.1061/(ASCE)0733-947X(2005)131:8(640). [11] Y. Cao, Z. Z. Yang, and Z. Y. Zuo, "The effect of curb parking on road capacity and traffic safety," European Transport Research Review, vol. 9, no. 1, pp. 1–10, Mar. 2017, https://doi.org/10.1007/s12544-016-0219- 3. [12] S. Biswas, S. Chandra, and I. Ghosh, "Side friction parameters and their influences on capacity of Indian undivided urban streets," International Journal of Transportation Science and Technology, vol. 10, no. 1, pp. 1– 19, Mar. 2021, https://doi.org/10.1016/j.ijtst.2020.03.007. [13] K. A. Madhu and H. N. Rajakumara, "Effect of Improper Bus Stop Locations on Capacity and Speed-Flow Relations on Urban Roads," International Journal of Recent Technology and Engineering, vol. 8, no. 3, pp. 1489–1494, Sep. 2019, https://doi.org/10.35940/ijrte.C4148. 098319. [14] A. M. Rao, S. Velmurugan, and K. M. V. N. Lakshmi, "Evaluation of Influence of Roadside Frictions on the Capacity of Roads in Delhi, India," Transportation Research Procedia, vol. 25, pp. 4771–4782, Jan. 2017, https://doi.org/10.1016/j.trpro.2017.05.489. [15] M. L. M. Chiguma, "Analysis of side friction impacts on urban roads links," Ph.D. dissertation, KTH Royal Institute of Technology, Stockholm, Sweden, 2007. [16] Nahry, H. R. Agah, A. Thohirin, and N. H. A. Hamid, "Modeling the Relationship between On-Street Parking Characteristics and through Traffic Delay: Parking and Traffic Delay Relationship," Proceedings of the Pakistan Academy of Sciences: A. Physical and Computational Sciences, vol. 56, no. 2, pp. 29–36, 2019. [17] I. Adinarayana and N. C. Anil, "The Study Exploration towards Side Friction Influences by traffic performance measures on roads," International Journal of Science Engineering and Advance Technology, vol. 5, no. 11, pp. 1024–1031, Dec. 2017. [18] K. L. Bang, "Impact of Side Friction On Speed-Flow Relationships for Rural and Urban Highways," SWEROAD, Indonesia, HDM4 project report, 1995. [19] K. F. Aronsson and K. L. Bang, "Influence on vehicle speed profiles of interactions with other road users," in European Transport Conference, Strasbourg , France, Sep. 2006, pp. 1–14. [20] B. R. Kadali, T. Chiranjeevi, and R. Rajesh, "Effect of pedestrians un- signalized mid-block crossing on vehicular speed," International journal for traffic and transport engineering, vol. 5, no. 2, pp. 170–183, 2015, https://doi.org/10.7708/ijtte.2015.5(2).07. [21] M. O. Kuttan, S. Babu, and G. Asaithambi, "Impact of Pedestrian Road Crossing on Capacity and Level of Service of Urban Undivided Roads in Indian Traffic Conditions," in Transportation Research Board 96th Annual Meeting, Washington, DC, USA, Jan. 2017, pp. 1–17. [22] A. Dhamaniya and S. Chandra, "Influence of Undesignated Pedestrian Crossings on Midblock Capacity of Urban Roads," Transportation Research Record, vol. 2461, no. 1, pp. 137–144, Jan. 2014, https://doi.org/10.3141/2461-17. [23] M. Angın and S. I. A. Ali, "Analysis of Factors Affecting Road Traffic Accidents in North Cyprus," Engineering, Technology & Applied Science Research, vol. 11, no. 6, pp. 7938–7943, Dec. 2021, https://doi.org/10.48084/etasr.4547. [24] S. Ranjan and V. Singh, "HEC-HMS based rainfall-runoff model for Punpun river basin," Water Practice and Technology, vol. 17, no. 5, pp. 986–1001, Apr. 2022, https://doi.org/10.2166/wpt.2022.033. [25] S. Ranjan and V. Singh, "ANN and GRNN-Based Coupled Model for Flood Inundation Mapping of the Punpun River Basin," Engineering, Technology & Applied Science Research, vol. 13, no. 1, pp. 9941–9946, Feb. 2023, https://doi.org/10.48084/etasr.5483. [26] A. Jawed, M. a. H. Talpur, I. A. Chandio, and P. N. Mahesar, "Impacts of In-Accessible and Poor Public Transportation System on Urban Enviroment: Evidence from Hyderabad, Pakistan," Engineering, Technology & Applied Science Research, vol. 9, no. 2, pp. 3896–3899, Apr. 2019, https://doi.org/10.48084/etasr.2482. [27] K. Srivastava and A. Kumar, "Critical Analysis of Road Side Friction on an Urban Arterial Road," Engineering, Technology & Applied Science Research, vol. 13, no. 2, pp. 10261–10269, Apr. 2023, https://doi.org/ 10.48084/etasr.5603.