Naghawi 2018
Naghawi 2018
Transportation Engineering
the Superstreet Unconventional
Intersection Design in Jordan
46(3), pp. 122-128, 2018
https://doi.org/10.3311/PPtr.11635
Creative Commons Attribution b Hana Naghawi1*, Alaa AlSoud2, Taqwa AlHadidi3
Abstract 1 Introduction
This paper provides policy makers with an objective assess- Transportation agencies and engineers around the world are
ment on the possibility for implementing an unconventional constantly challenged by the continuous increase in traffic vol-
arterial intersection design, the Superstreet, for improving umes and the resultant congestion at signalized arterial intersec-
safety and operational characteristics for an existing signal- tions. This traffic congestion is mainly caused by the high left
ized intersection. The signalized intersection selected for anal- turning traffic volumes, which inversely impacts performance at
ysis was Tabarbour, located in Amman, Jordan. Tabarbour signalized intersections. Transportation engineers have imple-
intersection is considered one of the most significant signal- mented many conventional measures to ease this problem, such
ized intersections in Amman. The intersection currently suffers as signal timing optimization, exclusive left turning signal,
from the following problems: long queues, delays and exces- implementing double left turn lanes, widening intersections, and
sive emissions. SYNCHRO microscopic simulation software grade separation, but the use of these conventional measures are
was used for signal optimization of the current conventional limited and expensive (Hummer and Reid, 1999). Therefore,
intersection. VISSIM microscopic simulation software was there has been considerable interest in alternative measures to
used to analyze and compare the Superstreet design to the con- improve performance and safety characteristics at signalized
ventional design. It was found that the proposed Superstreet arterial intersections with heavy left turning traffic. The Federal
reduced the average delay per vehicle by up to 87% and Highway Administration (FHWA, 2004) proposed several
reduced the maximum queue length by almost 97%. This unconventional arterial intersection designs (UAIDs). These
resulted in improving the level of service from F to C. Finally, UAIDs include Bowtie, Continuous Flow Intersection (CFI),
five year future traffic demand analysis was performed; it was Paired Intersection, Jughandle, Median U-Turn (MUT), Single
found that the Superstreet design didn’t perform as expected Quadrant Roadway, Superstreet Median Crossover, Continuous
under heavy traffic volume. Green T, and Parallel Flow Intersection (PFI). Typically, these
designs eliminate/reroute conflicting left turning manoeuvres to
Keywords and from the minor cross road.
Superstreet, Unconventional Arterial Intersection Design, UAID, Many unconventional intersection design studies are based
Microsimulation, VISSIM, SYNCHRO on theoretical traffic conditions. This leaves policy makers a
little worried of its effectiveness under real traffic conditions.
The purpose of this paper is to provide policy makers with an
objective assessment on reconfiguring an existing intersection
in Amman, Jordan into a Superstreet unconventional intersec-
tion design. The signalized intersection selected for analysis
was Tabarbour signalized intersection. This intersection is con-
sidered one of the most significant signalized intersections in
Amman. The intersection currently suffers from the following
1
Civil Engineering Department, University of Jordan, Amman 11942, Jordan problems: long queue of vehicles, delay and excessive emis-
2
Taffic Systems Management Division, Traffic Operation Department, Greater sions. The existing intersection was modelled using real traffic
Amman Municipality, Amman, Jordan data. For the purpose of this study the Microscopic simula-
3
Virginia| VT, Department of Civil and Environmental Engineering, tion softwares VISSIM and SYNCHRO were used to model
Blacksburg, VA, 24060, USA and compare the existing intersection to the reconfigured
*
Corresponding author, e-mail: hana_naghawi@yahoo.com Superstreet intersection. The following sections describe the
The Possibility for Implementing the Superstreet Unconventional Intersection ... 2018 46 3 123
conventional intersection design. Minseok et al. (2007) found Transportation engineers at Greater Amman Municipality
that the Superstreet intersection design can achieve significant (GAM) have implemented many conventional measures, to
reduction in accident frequency and severity, stopped delay ease the congestion problem at Tabarbour intersection, includ-
and queue length as compared to the conventional intersection ing: traffic signal coordination with adjacent signalized inter-
design. Also, Hochstein et al. (2009) analyzed crash data from sections on the arterial road, increasing the number of signal
Superstreet intersections in Maryland and North Carolina. One phases and implementing multiple left and right turning lanes.
site was converted from two way stopped controlled inter- Unfortunately, these conventional measures did not result in
section to a Superstreet intersection. It was found that the improving traffic performance at the intersection.
Superstreet intersection design reduced crash rate at the main
intersection by 92 percent. 4.2 Data Collection
Real traffic data were made available by the Department of
4 Methodology Traffic Operations at GAM. Traffic volumes including turning
4.1 Site Description movements, were collected at Tabarbour intersection on Monday
Fig. 2 shows an aerial photo of the selected intersection: the 27th of September 2014 at fifteen minute intervals. A sample
Tabarbour, Tabarbour intersection is located in Amman, the of the collected traffic volumes is shown in Table 1. It includes the
capital city of Jordan. The intersection is controlled by a pre- fifteen minute volumes in the evening between (7:30 - 9:30) pm.
timed traffic signal with a cycle length of 155 seconds. It is It can be seen that the major arterial, Al Shaheed, road suffers
considered one the most significant intersections in Amman. from high through traffic volumes, while the minor arterial roads,
The intersection connects the major arterial road: Al Shaheed, Prince Hamza and Tareq, suffer from high left turning volumes.
with the minor arterial roads: Prince Hamza and Tareq. Al
Shaheed road runs East and West bounds of the intersection. Table 1 Traffic Volumes at the Study Area (GAM)
It is a four lane divided arterial road with a posted speed of 70 App. Al Shaheed EB Al Shaheed WB
km/hr. It connects Zaraqa city, the second largest city in Jordan, Time
Mov. R T L U R T L U
with Amman. Also, the capital retail, commercial and industrial
7:30-7:45 95 350 81 7 48 580 97 1
facilities are located on the road to take advantage of its high
traffic volume and connectivity. The east side of the intersec- 7:45-8:00 110 379 84 6 54 495 113 14
tion is the entrance to a large residential area. The minor arterial 8:00-8:15 129 336 72 8 52 472 105 2
road is composed of Tareq road to the North and Prince Hamza 8:15-8:30 131 356 70 7 39 490 53 0
road to the South. Tareq road is a two lane divided minor arte-
Total 465 1421 307 28 193 2037 368 17
rial road with a posted speed of 60 km/hr. Prince Hamza road
8:30-8:45 117 380 35 6 55 576 59 0
is a three lane divided minor arterial road with also a posted
speed of 60 km/hr. 8:45-9:00 92 387 60 7 47 502 76 1
8:15-8:30 74 81 150 15 96 77 98 9
8:45-9:00 78 60 65 12 90 48 96 14
The intersection currently suffers from the following problems:
9:00-9:15 50 84 70 17 80 46 89 2
• Long queue of vehicles on all approaches.
• Enormous delays on all approaches. 9:15-9:30 75 94 75 10 60 60 67 8
• Large amount of emissions which are generated from Total 273 293 320 58 216 216 319 35
the acceleration and deceleration of vehicles at the
traffic signal.
Average Delay
Movement v/c LOS
(Sec/veh.)
N-S 1.14 149 F
N-E 1.13 163 F
N-W 0.26 0 A
S-N 2.42 672 F
S-W 1.66 357 F
S-E 0.35 1 A
W-E 0.81 51 D
W-N 1.09 137 F
W-S 0.35 1 A
E-W 1.17 124 F
E-S 1.42 237 F
E-N 0.17 0 A
The Possibility for Implementing the Superstreet Unconventional Intersection ... 2018 46 3 125
results side by side in an attempt to find the difference between 5.2 Disaggregated Level of Analysis
them. While the statistical validation uses measures as goodness Table 4 shows the comparison results of the VISSIM anal-
of fit, statistical tests, and the confidence intervals as validation ysis for the existing conventional four-legged signalized inter-
keys. These keys are used to quantify the similarity between section and the proposed Superstreet unconventional intersec-
observed and simulated values (Toledot and Koutsopoulos, tion disaggregated by movement using the average delay per
2004; Hummer and Jagannathan, 2008; Naghawi, 2012) vehicle (sec/veh.) as a LOS measure. It can be seen that most
The most popular goodness of fit measure is the root mean traffic movements experienced 13 to 87 percent reduction in
square percent error (RMSPE). The RMSPE is used to replicate the average delay per vehicle when using the Superstreet inter-
the error as a percentile rate (Daiheng et al., 2004). section design, except for the traffic coming from the North,
The RMSPE can be calculated using the following equation: Tareq minor arterial road, and heading to the East, Al Shaheed
major arterial road, on which the average delay per vehicle
Y −Y
2
1 N
RMSPE =
N
∑1 simY observed (1) was increased by 25 percent. Also, the Table shows that the
observed Superstreet intersection design significantly improved the LOS
(Daiheng et al., 2004) for all movements except for the North - East movement which
Where: remained the same as expected.
N is the number of simulation Runs.
Ysim is the simulation run throughput volume. Table 4 Average Delay per Vehicle for Conventional and Superstreet
Intersection Designs
Yobserved is the real throughput volume.
A total of thirty simulation runs using different random seed Conventional Superstreet
Reduction
numbers were executed. The average simulated throughput vol- Movement Avg. Delay Avg. Delay
LOS LOS (%)
ume on Al Shaheed road was found to be 5,920 vph, while the Sec/veh. Sec/veh.
observed throughput volume on Al Shaheed road was 5,652 vph. N-S 40.5 D 35.2 C -13.09
Based on Eq. (1) the RMSPE was found to be 0.87 %, N-E 38.3 D 47.9 D 25.07
this value is less than 15%, which is the RMSPE threshold N-W 4.5 A 0.6 A -86.67
(Hourdakis et al., 2002). This indicates that the model repli-
S-N 68.6 E 47.2 D -31.20
cates reality with high accuracy.
S-W 103.2 F 47.9 D -53.59
S-E 1.2 A 0.6 A -50.00
5 Operational Evaluation
In this paper, the aggregated results for the entire intersec- W-E 73.6 F 37.2 D -49.47
tion were compared first. Then the disaggregated results for W-N 244.5 F 34.5 C -85.89
each movement were computed and compared. W-S 18.3 B 10.4 A -43.17
E-W 112.9 F 17.8 B -84.23
5.1 Aggregated Level of Analysis E-S 290.3 F 37.2 D -87.19
Table 3 shows the aggregated results of VISSIM analysis
E-N 60.7 E 16.6 B -72.65
for the existing conventional four-legged signalized intersec-
tion and the proposed Superstreet unconventional intersection.
The results include the average delay per vehicle (sec/veh.), Table 5 shows a comparison of the maximum queue
LOS and the maximum queue length. These performance mea- length, disaggregated by movements, for the conventional
sures reflect traffic flow and discomfort levels caused to drivers and the Superstreet intersection designs. It can be seen that
on interrupted flow facilities. It can be seen that the intersec- the Superstreet unconventional intersection design reduced
tion delay per vehicle was reduced by 70.44% on the proposed the maximum queue length by 24 to 97 percent for almost all
Superstreet compared to the conventional intersection design, movements when compared to the conventional intersection
as a result the intersection LOS was improved from F (forced design except for the North - East traffic, on which the maxi-
conditions) to C (stable conditions). Also, the maximum queue mum queue length was increased by almost 40 percent. This is
length was reduced by almost 73% on the Superstreet when consistent with the high average delay per vehicle for the North
compared to the conventional intersection design. - East turning movement.
The Possibility for Implementing the Superstreet Unconventional Intersection ... 2018 46 3 127
Federal Highway Administration. (2014). Restricted Crossing U-turn Minseok, K., Chang, G., Rahwanji, S. (2007). Unconventional Arterial
Informational Guide, Publication FHWA-SA-14-070. [Online]. Intersection Designs Initiatives. In: IEEE Conference on Intelligent
Available from: https://safety.fhwa.dot.gov/intersection/alter_design/ Transportation Systems, Seattle, Washington.
pdf/fhwasa14070_rcut_infoguide.pdf [Accessed: 27th March 2016]. Jae, M., Kim, Y., Kim, D., Lee, S. (2011). The Potential to Implement a
Hochstein, J., Maze, T., Welch, T., Preston, H., Storm, R. (2009). The J-Turn Superstreet as an Unconventional Arterial Intersection Design in Korea.
Intersection: Design Guidance and Safety Experience. In: Transportation KSCE Journal of Civil Engineering. 15(6), pp. 1109–1114.
Research Board 88th Annual Meeting, Washington, D.C. https://doi.org/10.1007/s12205-011-1157-1
Hourdakis, J., Michalopoulos, P., Kottommannil, J. (2003). Practical Procedure Municipality of Greater Amman. (2014). Traffic Operation Department reports.
for Calibration Microscopic Traffic Simulation Models. Transportation Naghawi, H., Idewu, W. (2014). Analysing Delay and Queue Length Using
Research Record: Journal of the Transportation Research Board. 4167, pp. Microscopic Simulation for Unconventional Intersection Design
130-139. Superstree. Journal of the South African Institution of Civil Engineers.
https://doi.org/10.3141/1852-17 56(1), pp. 100-107.
Hummer, J., Jagannathan, R. (2008). An Updates on Superstreet Implementation Naghawi, H. (2012). Methodological Notes on Regional level Validation
and Research. In: 8th National Conference on Access Management, of Microscopic Traffic Simulation Model. Jordan Journal of Civil
Transportation Research Board, Baltimore, MD. Engineering. 6(2).
Hummer, J., Reid, J. (1999). Unconventional Left-Turn Alternatives for Urban Reid, J. (2004). Unconventional Arterial Intersection Designs, Management
and Suburban Arterials- An Update. In: Transportation Research Circular and Operational Strategies. Persons Brinckerhoff, New York, USA,
E-C019: Urban Street Symposium Confirence Proceedings, pp. 28-30. Reid, J., Hummer, J. (1999). Analyzing System Travel Time in Arterial
Dallas, TX. Corridors with Unconventional Designs Using Microscopic Simulation.
Hummer, J., Schroeder, B., Moon, J., Jagannathan, R. (2007). Recent Superstreet Transportation Research Record: Journal of the Transportation Research
Implementation and Research. In: 3rd Urban Street Symposium: Uptown, Board. 1678, pp. 208-215.
Downtown, or Small town: Designing Urban Street that Work, June https://doi.org/10.3141/1678-25
24-27, Seatlle, Washington, Transportation Research Board of National Toledot, T., Koutsopoulos, H. (2004). Statistical Validation of Traffic
Academies, Washington, D.C. Simulation Models. Transportation Research Record: Journal of the
Kim, T., Edara, P. K., Bared, J. (2007). Operational and Safety Performance Transportation Research Board. 1876, pp. 142-150.
of a Non-Traditional Intersection Design: The Superstreet. In: https://doi.org/10.3141/1876-15
Transportation Research Board 86th Annual Meeting, Compendium of
Papers CD-ROM, TRB, Washington, DC.