Improvement of pedestrian safety and traffic flow regulation at the railway junction, Kandy

The junction near the Kandy railway station needs improvement as a solution to the traffic congestion at the location, especially during the peak hours of the day. The existing problem was studied in detail and several reasons for this problem were identified. To identify the movement of traffic and pedestrians at the railway junction, traffic studies were conducted during weekdays and weekends. Vehicular traffic by vehicle category, vehicle movements, and pedestrian traffic patterns were studied after identifying the three peak hours (morning, mid-day and evening) of the day. A heavy pedestrian movement surrounding the study locality was observed due to the presence of several main government institutions around that junction. Pedestrian traffic flow, pedestrian volumes and the level of service were calculated by the results collected from the pedestrian traffic surveys. A questionnaire survey was carried out during the peak hours to collect the views of the pedestrians about the existing situation at the location and the results were used in the analysis. As a measure to reduce the traffic congestion and improved vehicular flow and pedestrian safety, an underground pedestrian crossing structure together with an at-grade signalised scheme was proposed and designed.


INTRODUCTION
The Kandy railway junction is surrounded by several main government institutions.At present there are three pedestrian crossings across each road, a few meters away from the junction (Figure 1) with a heavy pedestrian movement (Appendices A and B), interrupting the vehicular flow at the intersection.According to past accident data gathered from the Kandy Traffic Police (2010) fatal and serious accidents have occurred around the vicinity due to heavy pedestrian activities in the area.
This study was conducted to address the problem of traffic congestion with the view of streamlining the vehicular flows and improving pedestrian safety at the location by minimising the interaction between pedestrians and vehicular movements.The main objective of this study was to examine the existing situation and provide a solution to facilitate safe and comfortable means of road crossing for pedestrians while allowing the traffic to flow smoothly.

METHODOLOGY
Several traffic surveys were conducted during the weekdays and weekends to identify the three peak hours of the day, peak hour traffic flows and turning movements of different vehicle categories at the railway junction.In addition, the pedestrian traffic surveys were also conducted to identify the peak hours and volumes of pedestrians using the three pedestrian crossings located in each road during the peak hours.

Identification of peak hours
The traffic flow of a road does not remain constant throughout the day or week, but varies both in space and time.A 'peak hour' represents the most critical period for operations and has the highest demand requirements for a given location.Therefore, capacity and other traffic analysis has to focus on the peak hour flows.
The Kandy town has a 5 % annual traffic growth rate (RDA, 2000).A major base road (William Gopallawa Mawatha) was selected to carryout the traffic survey on a weekday from 6:30 am to 9:30 am, 12:00 noon to 2:00 pm, and 4:30 pm to 6:30 pm, and on weekends from 7:30 am to10:30 am, 12:00 noon to 2:00 pm, and 4:00 pm to 6:00 pm, when the traffic is at its heaviest in order to identify the three peak hours of the day during morning, mid-day, and evening respectively.The counts were recorded at 15 min intervals.From the traffic surveys, the peak hours and the respective two-way traffic volumes were identified (Table 1).

Classification of vehicles
The vehicles were categorized into ten separate groups.Passenger car unit (PCU) is used for expressing various types of vehicles having different characteristics by a common equivalent unit.PCU factors for vehicles in flat terrain were considered due to the nature of the site and are listed in Table 2.

Turning movement survey
Turning movement surveys were carried out during the identified peak hours on a weekday and weekend, for the ten vehicle categories at the railway junction, which is a channelized intersection.All the roads have two lanes in each direction and it had a mixed flow pattern consisting of crossing, diverging and merging movements.Diverging and merging from the left side did not cause much problem but, crossing and diverging to and merging from the right side caused conflicts and difficulties to traffic moving on a straight path.The volume of vehicles making each turning movement was not constant during all three peak hours.It changed from one period to another and also within the peak hour.At present, traffic at the intersection is controlled by a policeman to minimise congestion during peak hours.

Pedestrian survey
Pedestrian surveys were carried out during weekdays and weekends to observe the volume of pedestrians using the three pedestrian crossings located in each road a few meters away from the junction for 12 consecutive 5 min periods, for each of the three peak hours (Appendices A and B).The rate of flow did not remain constant and crossing at Goodshed end, as most of the government buildings and the main bus stand are located on that part of the junction.
varied within the hour, and within the week.A high pedestrian volume was observed during the weekends.It was observed that most of the pedestrians used the

Accident details
Information about the accidents for the past ten years at the railway junction and near the clock tower junction was collected from the Kandy Police Station (2010).Types of accidents that occurred at both places were property damages, injuries, serious injuries and fatal accidents.Most of the accidents had occurred due to the carelessness of both the pedestrians and the drivers.
After the construction of the underground pedestrian crossing structure near the clock tower junction at Kandy, a vast improvement of vehicular and pedestrian traffic flow and pedestrian safety at the locality was observed (Kandy Police station, 2010).There is a heavy vehicular traffic as five roads meet at this particular junction.Before the construction of the underground pedestrian crossing structure, there were six pedestrian crossings in the near vicinity and the police used to stop vehicles for long durations, which created traffic congestion at the clock tower junction.Currently the sidewalks are provided with guard rails to prohibit people crossing the road at unauthorized locations.All the pedestrian crossings in the vicinity are also closed.As the pedestrians are using the pedestrian underpass, the traffic congestion and the number of accidents at the clock tower junction have decreased considerably (Figure 2).Two fatal accidents were recorded during the construction stage of the pedestrian underpass at the clock tower junction.The problem at the Kandy railway junction is comparable to the clock tower junction.It is seen from the accident data in Table 4, that four fatal accidents had taken place during the past ten years at the Kandy railway junction.It should be noted that the real number of 'property damage' accidents is more than that is indicated in Table 4 from 2006 to 2010 due to the prevailing 'on the spot settlements' during this period.Providing a pedestrian crossing structure at the Kandy railway junction and preventing the road crossing other than using the pedestrian crossing structure appears to be a reasonable solution to the above problems.

Questionnaire survey
A questionnaire survey was carried out during the three peak hours to collect the views of the pedestrians, about the existing crossing conditions at the Kandy railway junction and to obtain their suggestions to improve it.A sample of pedestrians between the age of 15 -70, who were passing the railway junction for various purposes was selected for the questionnaire survey.The results indicated that the pedestrians were not satisfied with the existing pedestrian crossings at the particular junction because of the following reasons, • The existing pedestrian crossings are not properly located, • It takes more time to cross the road by using the pedestrian crossings because of the heavy traffic at the junction, • Drivers of the vehicles do not like to stop their vehicles at the pedestrian crossings, • The feel highly insecure existing pedestrian crossings The pedestrians always wish to cross the road safely, and their suggestions to improve the pedestrian traffic flow at the junction with enhanced safety were obtained.Suggestions by pedestrians to improve the vehicular flow, pedestrian traffic flow, and safety are shown in Figure 3.
It was also noted that all the pedestrians that answered the questionnaire had already used the existing pedestrian underpass at the clock tower junction.Among those who were interviewed 93 % suggested constructing an underpass at the Kandy railway junction, based on the convenience they had experienced at the nearby underpass at the clock tower junction.
Journal of the National Science Foundation of Sri Lanka 41(3) September 2013 Suggestions obtained through the survey were: • To install direction boards to identify the correct way that should be selected after entering the pedestrian underpass, specially for the use of those who are not residents of Kandy town • There should be a good illumination system for both day and night to ensure the security of the pedestrians • There should be planned precautions for emergency situations (fire, flood, heavy rain etc.) • Control the entry of unwanted people into the pedestrian underpass • Facilities for the elderly and disabled pedestrians should be provided According to the above results it could be concluded that the most of the pedestrians favoured a pedestrian underpass for the Kandy railway junction, as a solution for the existing inconvenience at the location.

Capacity analysis and design of roads
From the data collected through traffic surveys, the existing traffic flow rates and the peak hours were identified.The corresponding Levels of Services (LOS) during the peak hours were calculated and the road was improved and designed to cater to the projected traffic 20 years from the present.

Peak hour factor (PHF)
The maximum hourly volume of the day divided by the rate of flow during peak 15 minute period within the Peak Hour is a measure of traffic demand fluctuation within the Peak Hour.
Service flow (SF) The actual rate of flow for the peak 15 minute period was expanded to an hourly volume and expressed as vehicles per hour or vehicles per hour per lane.

Multilane highways
Capacity analysis and design was carried out along the William Gopallawa Mawatha and at two ends of the S.W.R.D. Bandaranayake Mawatha as recommended by the US Highway Capacity Manual (1985).
Where, The details of existing and designed/proposed roads are indicated in Table 5.

Traffic signal design for Kandy railway junction
The critical volume of vehicles approaching towards the railway junction during the morning peak is shown in Figure 4.
After considering all the different traffic flow scenarios the following stage arrangement was established to facilitate all the traffic flow movements with a minimum number of stages (i.e., total of 3 stages as indicated in Figure 5).Since there is no all-red phase, R = 0 There are three phases and the total loss time (L) can be calculated Actual green G ai each phase is obtained by, ‫ܩ‬ ൌ ‫ܩ‬ ݈ െ ߬

Capacity analysis and design of pedestrian crossing structure
From the data collected from the pedestrian traffic surveys (Appendices A and B), the existing pedestrian flow rates during the peak hours and the Levels of Service (LOS) were calculated.Then the pedestrian crossing structure was designed to cater to the pedestrian volumes projected for 20 years.The basic relationships for pedestrian design (Transportation Research Board, 1985).Similarly the subway widths of all three crossings were calculated (Table 6).
It should be noted that when planning for pedestrians the main objective should be to encourage orderly and safe movement of pedestrians and vehicles without interference from each other.Pedestrians should be guided away from traffic danger spots in main roads by providing proper facilities, and without interference from the moving traffic as much as possible (Weerasekera, 2009).

Pedestrian crossing structure
The pedestrian crossing structure consists of three accesses tunnels, which are from the police quarters

Figure 2 :
Figure 2: Road accidents at clock tower junction (before and after the construction of pedestrian underpass) flow rate; the maximum flow rate that can be accommodated by the multilane highway segment under study in one direction under prevailing roadway and traffic conditions, while meeting the performance criteria of LOS i , in vph c j = capacity per lane for a multilane highway with design speed j N = number of lanes in one direction (v/c) j = maximum volume-to-capacity ratio while maintaining the performance characteristics of LOS j f w = adjustment factor for lane width and/or lateral clearance restrictions f HV = adjustment factor for the presence of heavy vehicles in the traffic stream f E = adjustment factor for the development environment and type of multilane highway f p = adjustment factor for driver population E T , E B = passenger car equivalents for trucks and buses, respectively P T , P B = proportion of trucks and buses, respectively in the traffic stream September 2013Journal of the National Science Foundation of Sri Lanka 41(3)By using Webster's method(Salter & Hounsell, 1996):Determination of equivalent hourly flow: lost time for phase i R = total all -red time during the cycle

Figure 4 :ܻ
Figure 4: The critical volume of vehicles approaching towards the railway junction during the morning peak

Figure 5 :
Figure 5: Staging diagrams for traffic flow movements

Figure 6 :
Figure 6: Phasing diagrams for the Kandy railway junction

Forecasting the pedestrian flow
in n years, P n = P 0 (1 + r / 100) n Where, P n = forecasted pedestrian flow for n years P 0 = present pedestrian flow r = reasonable growth rate factor n = number of years Designing the width of the subway for an assumed LOS Specimen calculation (William Gopallawa Mawatha) Total volume crossing (peak hour) = 650 ped/hr Pedestrian flow rate per unit per unit width of crossing = 650/60 ped/min/ft = 11 ped/min/ft As per the Highway Capacity Manual (1985) the existing Level of Service is D. Hence, forecasted pedestrian flow for 20 years, P 20 = 650 (1+ 11/100) 20 = 810 ped/hr Design width of subway for LOS B, Assume; Pedestrian flow rate = 2.5 ped/min/ft ‫ܩ‬ ൌ ‫ܩ‬ ݈ െ ߬ ൌ Proposed width = 2 x 810/(60 x 2.5) = 10.8 ft ( 3.3 m) Hence proposed subway width = 4.0 m

Table 2 :
Equivalent passenger car units (PCU) for flat terrain and two-way roads

Table 3 :
Identified peak hours for weekdays and weekends

Table 4 :
Accidents at the Kandy railway junction

Table 5 :
Results of the existing and designed roads

Table 5 : Results of existing and Figure 7: Proposed
pedestrian subway at Kandy Railway Junction