Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

The Effect of the Number of Right-Turn and Left-Turn Lanes on the Performance of Undersaturated Signalized Intersections

The Effect of the Number of Right-Turn and Left-Turn Lanes on the Performance of Undersaturated... Hindawi Journal of Advanced Transportation Volume 2023, Article ID 8764498, 12 pages https://doi.org/10.1155/2023/8764498 Research Article TheEffectoftheNumberofRight-TurnandLeft-TurnLanesonthe Performance of Undersaturated Signalized Intersections 1 2 1 Omid Rahmani , Amir Saman Abdollahzadeh Nasiri , and Iman Aghayan Faculty of Civil Engineering, Shahrood University of Technology, Shahrood, Iran Department of Civil Engineering, South Tehran Branch, Islamic Azad University, Tehran, Iran Correspondence should be addressed to Omid Rahmani; omid.rahmani@shahroodut.ac.ir Received 6 August 2022; Revised 17 October 2022; Accepted 23 January 2023; Published 16 February 2023 Academic Editor: David Rey Copyright © 2023 Omid Rahmani et al. Tis is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Vehicle trafc fow channelizing can signifcantly contribute to fewer collisions and smooth trafc fow in isolated signalized intersection design. Terefore, this study investigates the efect of changes in the number of right-turn lanes with splitter islands and left-turn storage bays for diferent volume approaches on intersection performance. Tis study aims to provide an accurate analysis to estimate the efect of such changes on signalized intersections, which can improve intersection performance. Furthermore, the delay parameter derived from HCM2016 and simulation was evaluated for diferent scenarios. For this purpose, simulation was performed using Synchro software. In this study, a symmetrical, 90-degree, undersaturated (volume per capacity lower than 1, V/C < 1), and four-leg intersection were considered according to AASHTO Green Book 2018 suggestions. HCM2016 results indicated the delay parameter was less sensitive to the undersaturated condition than the right- turn volume variation for both single and dual-lane bays. However, the simulation results indicated that the delay parameter was not constant, depending on the number of right-turn lanes and volumes. On the other hand, there was a signifcant diference in delay parameters between the single and dual left-turn lanes for both simulation and HCM cases estimated 77.4% and 59.7%, respectively. Overall, this study can provide a vision for trafc engineers to modify the geometry of the four-leg signalized intersections, if the right-turn or left-turn demand volume of an undersaturated signalized intersection is larger than the through approach. of the intersection is considered when it is not feasible to 1. Introduction reduce the delay time of a heavy trafc intersection through Te intersection capacity plays an important role in de- trafc light timing and phasing. Right-turn splitter islands termining the trafc capacity of an urban road network. and left-turn storage bays can be alternative geometric Regarding their heavy trafc, signalized intersections have characteristics to improve the signalized intersections. Tese been the focus of many studies because a large portion of a techniques can signifcantly reduce collisions and contribute to smooth trafc fow [4, 5]. vehicle’s lost time is related to signalized intersections. Te main methods to reduce delays and improve the intersection Trafc fow has been less considered as a statistical capacity are to optimize the trafc light timing and modify criterion to change left- and right-turn volumes [6]. If ac- the geometric characteristics of the intersections [1]. curate predictions can be made to estimate the efect of Safety and capacity play the most important role in the geometric modifcations on left and right turns, a better geometric design of an isolated signalized intersection [2, 3]. vision can be outlined for the design of signalized inter- Trafc signal timing and phasing optimization without sections before saturation. Te geometric parameters for the changing the geometric characteristics of the intersection are design of the left and right turns in a signalized intersection possible to limited values. Terefore, geometric modifcation include bay width, channelized right-turn radius, right-turn 2 Journal of Advanced Transportation lane length, left-turn storage length, taper length, and the undersaturated isolated signalized intersection considering number of lanes [7]. Most studies have considered right-turn the geometric characteristics (i.e., number of lanes) of channelized right-turn approaches and the storage length of lane length and left-turn storage length, while the increased number of lanes and considering the storage or deceleration the left-turn approach. length can prevent lane blockage for incremental turning A summary of some previous studies related to signal- volume of the intersections [8, 9]. ized intersections is shown in Table 1. Based on the Te majority of studies have been dedicated to over- movement in diferent signalized intersection approaches, saturated (i.e., V/C≥ 1) or saturated signalized intersections the intersection assessment method (probability, simulation, [10]. Signalized intersection phasing and timing can be a or both), trafc fow parameters (queue length, delay, and V/ complex task in some cases due to the saturated incoming C), geometric parameters, and also channelized right-turn intersection trafc [11]. It is necessary to avoid saturation as movement were reviewed. Te HCM-based studies are also much as possible while designing and selecting the geo- presented. Tese studies propose a model or process to metric alternatives of an intersection. complement the HCM methodology or compensate for its weaknesses in signalized intersections. Te literature shown According to the Highway Capacity Manual (HCM) 2016 regulation, the delay is the main criterion (measure of in Table 1 is categorized based on the movement, including efectiveness (MOE)) for at-grade intersection efciency (1) left-turn lane, (2) right-turn lane, (3) right turn, left turn, evaluation [12]. Based on HCM2016, the delay calculation and through, and (4) right turn, left turn, through, and methodology for undersaturated signalized intersections is U-turn. less sensitive to changing the number of channelized right- Table 1 indicates that the geometric parameter evalua- turn lanes. In other words, the delay remained constant for tions aimed to improve the capacity of signalized inter- right-turn overdemand and fxed through approach volume. sections were focused on left- and right-turn storage bays. Also, the extent to which the changing number of bays Te simultaneous efect of geometric and trafc parameter afects the intersection delay is unknown. variations, including the number of right- and left-turn lanes Terefore, this study seeks to evaluate the efect of the and their trafc demand volume, was not included. Tese studies were carried out for near-saturated or oversaturated right-turn lane and left-turn storage bay quantity on the capacity of an undersaturated signalized intersection. To this signalized intersections. If the left- and right-turn volume of end, a symmetrical 90-degree four-leg signalized intersec- a signalized intersection increases, it is important to rec- tion without longitudinal grade was designed based on the ognize its limitations and maintain the intersection in an AASHTO guideline [13]. Te efect of the number of lanes undersaturated state. on the delay parameter was compared to the HCM2016 Te frst novelty of this study is to evaluate the efect of model. Te trafc and geometric characteristics of the un- the number of left- and right-turn lanes with storage length dersaturated isolated signalized intersection were modeled on the undersaturated signalized intersection capacity. Te and analyzed by Synchro11 software. Te analyses were assessment was performed for trafc volume scenarios of based on the delay and volume per capacity (V/C) ratio. variant left- and right-turn approaches and constant through approach volume. Tus, the combined efect of the number of lanes and trafc fow (left- and right-turn volume) was 2. Literature Review investigated. HCM assumes a constant delay for the un- Previous studies have focused on evaluating the perfor- saturated state up to a certain right-turn volume. Terefore, mance of signalized intersections with left- and right-turn another novelty of the study is to evaluate the delay and V/C storage bays and channelized right-turn lanes in terms of variations as the main criteria to determine the signalized delay, V/C, and queue length. In general, the research re- intersection capacity relative to HCM. garding the capacity of signalized intersections with storage bays can be divided into two categories: (i) the phasing and 3. Methodology timing of signalized intersections and (ii) the efect of geometric characteristics on the capacity of signalized Synchro trafc simulator was employed to defne the geo- intersections. metric characteristics, trafc volumes, trafc light timing, For the frst category, it was assumed that the capacity of and delay. Many studies regarding trafc in at-grade sig- an isolated signalized intersection was improved by phasing nalized intersections used Synchro [35–40]. By considering and timing optimization [11, 14–16]. In the second category, diferent HCM criteria [41, 42], Synchro was selected be- the intersection throughput was improved by changing the cause it contains the sixth edition of HCM (HCM2016). Te geometric characteristics of the left and right turns, such as average control delay (d) for each lane is based on the right-turn channelization and increased left- or right-turn following formula: storage length [17–20]. Te current study investigates an 􏽳���������������� � 3600 (3600/c)x ⎡ ⎢ ⎤ ⎥ ⎢ ⎥ ⎣ ⎦ (1) d � + 900 T (x − 1) + (x − 1) + + 5 × min[x, 1], c 450 T Journal of Advanced Transportation 3 Table 1: Studies related to signalized intersections. Evaluation methods Trafc parameters Movement HCM Proposed Geometric Right-turn Study (year) type checked model parameters channelized Simulation Probability Queue Delay v/c Kikuchi et al. Length of left- — — ✓ ✓ — — — — (2004) [2] turn lane Qi et al. (2007) Length of left- — — ✓ ✓ — — ✓ — [21] turn lane Yin et al. (2010) Length of left- ✓ CORSIM ✓ ✓ — — ✓ — [22] turn lane Kikuchi and Monte Carlo Length of left- Kronprasert Left-turn lane — method and ✓ ✓ — — ✓ — turn lane (2010) [23] VISSIM Reynolds et al. VISSIM Length of left- ✓ — ✓ — ✓ ✓ — (2010) [24] DYNASMART turn lane Guo et al. Length of left- — — ✓ — — ✓ ✓ — (2011) [8] turn lane Yin et al. (2011) Length of left- — VISSIM ✓ ✓ ✓ — ✓ — [25] turn lane HSC+ Qi et al. (2012) Synchro Length of left- ✓ — ✓ — — — — [26] VISSIM turn lane SimTrafc Yao and Length of left- Michael Zhang ✓ VISSIM — ✓ ✓ — ✓ — turn lane (2013) [11] Length of left- Yao (2016) [27] ✓ — ✓ — ✓ ✓ ✓ — turn lane Number of Ma et al. (2017) left-turn lane ✓ VISSIM ✓ — — ✓ ✓ — [28] in waiting area Liu et al. (2018) Length of left- ✓ VISSIM ✓ — — ✓ ✓ — [3] turn lane Zheng et al. Length of left- ✓ VISSIM — ✓ ✓ — ✓ — (2020) [18] turn lane Guo et al. Length of left- — VISSIM — ✓ — — — — (2021) [15] turn lane Kikuchi & Length of Kronprasert — — ✓ ✓ — — — right-turn — (2008) [29] lane Macfarlane Synchro et al. (2011) ✓ — — ✓ — ✓ — — (SimTrafc) [30] Chen et al. ✓ VISSIM ✓ — ✓ ✓ ✓ — — (2012) [31] Chen et al. Right-turn ✓ — ✓ — — — ✓ — ✓ (2013) [32] lane Length of Farivar et al. ✓ VISSIM ✓ — ✓ — ✓ right-turn ✓ (2016) [9] lane Length of Farivar et al. ✓ VISSIM ✓ — — ✓ ✓ right-turn ✓ (2017) [9] lane Li et al. (2019) ✓ VISSIM — — — ✓ ✓ — ✓ [33] Length of Kikuchi et al. ✓ VISSIM ✓ ✓ — — — right-turn and ✓ (2007) [34] left-turn lane Right turn, Li and left turn, and Elefteriadou ✓ CORSIM — ✓ ✓ — — — — through (2013) [17] Shatnawi et al. ✓ VISSIM — ✓ ✓ — — — — (2018) [19] 4 Journal of Advanced Transportation Intersection geometry Traffic demand Signalized intersection Intersection angle = 90° Traffic volumes: (Without longitudinal grades) Through= 400 (Veh/h) Single Right-turn lane = Initial signal cycle length 100 to 800 (Veh/h) Storage length for right-turn Signal phasing arrangement Dual Right-Turn Lanes = and left-turn (Based on (two-phases) 200 to 1450 (Veh/h) AASHTO 2018) Single Left-turn Lane = 100 to 400 (Veh/h) Single and dual lane for right Dual Left-turn Lane = 100 and left-turn to 700 (Veh/h) Right-turn curb radius (15.3 m) Synchro (Simtraffic) Lane width (3.65 m) Optimize signal cycle Critical gap = 5.0 (s) and length for each scenario Follow-up gap = 2.2 (s) (Based on AASHTO 2018) Simulation HCM 2016 Comparing outputs analysis outputs between Control (V/C) <1 single and dual- Delay (s) lane Delay (s) Control all demand volume (V/C) change scenarios for under saturation Comparing outputs between simulation and HCM Figure 1: Research method in detail. where x � volume-to-capacity ratio of each lane, c � capacity Figure 1 summarizes the defned inputs and outputs for of the subject lane (veh/h), and T � time period (h) � 0.25 simulation and the methodology details. Te drivers’ be- (h). havior in terms of gaps was selected based on AASHTO 2018 Te control delay for an approach is calculated using the selections. To this end, critical gaps and follow-up gaps were weighted average of the delay for each lane on the approach, accounted for in the simulations. Te results were used to weighted by the volume in each lane. Te calculation is compare the delay obtained by the simulation and HCM. demonstrated in Te volume per capacity (V/C) obtained by the simulation was evaluated for each scenario to control whether the in- 􏽐 d v i i d � , (2) coming right- and left-turn volumes were in an unsaturated 􏽐 v state (V/C< 1). A single and dual-lane method was adopted to compare the right- and left-turn movements. where d � control delay for the approach (s/veh), d � a i control delay for lane i (s/veh), and v � fow rate for lane i (veh/h). 3.1. Geometric Design of Intersections. For all designed Te control delay for the intersection as a whole is intersections, the intersections were 90 degrees without a similarly calculated by computing a weighted average of the longitudinal slope. Te radius of the channelized right- delay for each approach and weighted by the volume on each turn horizontal arc edge was 15.3 m. Te ideal lane width approach which is shown in the following equation: was 3.65 m. Te length of the left-turn storage bay was 􏽐 d v based on NCHRP 780 [43], which is suggested by the a a d � , (3) intersection latest version of the AASHTO Green Book (2018). To this 􏽐 v end, the left-turn storage bay was considered 30.5 m long where d � control delay for the entire intersection intersection because the through movement volume of the opposite (s/veh), d � control delay for approach a (s/veh), and v � a a trafc fow was 400 Veh/h. However, since 2% of the total fow rate for approach a (veh/h). incoming trafc was associated with heavy vehicles, 7.6 m Journal of Advanced Transportation 5 Right-Turn Channelized Island (Splitter Island) Storage Length Right-Turn Lane Storage (deceleration) Curb Radius Length Left-Turn Lane (Reserved Lane) Length of Taper (a) (b) Figure 2: Geometric design elements for turning movement at an intersection: (a) single right-turn lane and (b) single left-turn lane. was added to the left-turn storage bay length, according 4. Results’ Analysis and Discussion to AASHTO 2018, leading to a 38.1 m long left-turn After the scenario simulation, the results were analyzed and storage bay. compared. Te results were estimated for a single right-turn Figures 2(a) and 2(b) show the length of the right-turn lane, a single left-turn lane, a dual right-turn lane, and a dual bays and left-turn storage length after the geometric mod- left-turn lane. In addition, the results were presented in the ifcations. All the scenarios were considered in separate form of volume-delay control and V/C curves to analyze the phasing. First, the trafc light optimization was carried out intersection performance and compare diferent scenarios. by optimizing the intersection cycle length, green light Te scenarios were simulated for unsaturated trafc using timing for each route, and approaches. Te optimization was Synchro, and the results were compared to HCM2016. Tus, performed by the Synchro simulator. Ten, the total delay of the following results were obtained. the intersection was evaluated. Te undersaturated state (V/ C< 1) was the criterion for selecting the defned light or heavy trafc volumes. In addition to one right- and left-turn lane, dual right- and left-turn lanes were also addressed in 4.1. Single Right-Turn Lane. Te through and left-turn the current study. volumes were assumed constant for all directions (i.e., 400 and 150 Veh/h). Te right-turn volume varied in the range of 100–800 Veh/h with an incremental step of 50 Veh/h. Te 3.2. Trafc Demand Characteristics at the Intersection. simulation results showed that the delay parameter initially Two general volumetric cases with V/C< 1 were considered decreased for the trafc volume up to 300 Veh/h (Figure 5). to defne the scenarios. In the frst case, the through and left- For higher volumes, the delay parameter increased. turn approach volumes were fxed (400 and 150 Veh/h, HCM2016 result indicated that the delay was independent of respectively). Te volume growth rate was defned as 50 Veh/ the incoming volume below 500 Veh/h. As the volume in- h for each new scenario for right-turn movements. Te creased from 500 to 600 Veh/h, the delay increased, and for single right-turn and dual right-turn lane volumes were in higher volumes, the amount of delay increased signifcantly. the range of 100–800 and 200–1, 450 Veh/h, respectively. In For the trafc volume of 600 Veh/h, the delay reported by the second case, the through and right-turn approach vol- our simulations and HCM2016 was equal. For over 600 umes were fxed (400 and 150 Veh/h, respectively). Te (Veh/h), the delay of both HCM and simulation increases volume growth rate was defned as 50 Veh/h for each new because the delay of the single right-turn lane is sensitive to scenario for left-turn movements. Te single left-turn lane the incoming volume increase. In fact, the volume of 600 Veh/ and dual left-turn lane volumes were in the range of 50–400 h was the threshold for a sudden change in delay. As the and 200–1, 450 Veh/h, respectively (Table 2). After applying volume increased from 600 to 800 Veh/h, the growing trend the passing volume, the trafc light cycle length was opti- of the delay in the simulation was more drastic than that of mized by Synchro. Figure 3 shows the intersection plans for HCM2016, indicating that HCM2016 is a more conservative dual left-turn lanes (3a), dual right-turn lanes (3b), and method for delay estimation. For instance, the delay pa- single left- and right-turn lanes (3c). Figure 4 depicts the 3D rameter obtained by simulations was 11% higher at 650 Veh/h, view of the simulation of a scenario. while the diference was 26.7% at the volume of 800 Veh/h. 6 Journal of Advanced Transportation 200 200 400 400 150 150 (a) (b) (c) Figure 3: Sample of simulated scenarios in Synchro: (a) dual left-turn lanes, (b) dual right-turn lanes, and (c) single right- and left-turn lanes. Figure 4: Simulation of intersection in Synchro (3D). Table 2: Summary of scenarios, trafc demand volume, and outputs. Type of Number of Trough trafc Left-turn trafc volume Right-turn trafc volume (Veh/h) Outputs turn lanes volume (Veh/h) (Veh/h) 100, 150, 200, 250, 300, 350, 400, 450, 500, 550, Single lane 400 150 600, 650, 700, 750, 800 Right- 200, 250, 300, 350, 400, 450, 500, 550, 600, 650, turn Dual lane 400 700, 750, 800, 850, 900, 950, 1000, 1050, 1100, 150 Delay and 1150, 1200, 1250, 1300, 1350, 1400, 1450 volume to 50, 100, 150, 200, 250, Single lane 400 150 capacity (V/C) 300, 350, 400 Left- 100, 150, 200, 250, 300, turn Dual lane 400 150 350, 400, 450, 500, 550, 600, 650, 700 In addition, the diference between the simulation and lane. According to Table 3, the results demonstrate that there is HCM2016 control delay results was verifed statistically using a a signifcant diference between simulation and HCM2016 in two-sample F-test for variances method in a single right-turn control delay results at a 95% confdence level (P value <0.05). 150 Journal of Advanced Transportation 7 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800 850 Volume (Veh/h) simulation HCM 2016 Figure 5: Te efect of volume variations on delay for a single right-turn lane. obtained by simulation was 7.1% higher at the volume of 1,250 Veh/h, while the diference was 23.2% at the volume of Table 3: F-test performed between simulation and HCM2016. 1,450 Veh/h. Method Variance F P value F critical As mentioned in section 4.1, the diference between the Simulation 21.71 simulation and HCM2016 control delay results was verifed 5.99 0.001 2.48 HCM2016 3.62 statistically using a two-sample F-test for variances method in dual right-turn lane. Table 4 shows a signifcant diference between simulation and HCM2016 in control delay results at 4.2. Single Left-Turn Lane. Te through and right-turn a 95% confdence level (P value <0.05). volumes were considered constant for all directions (i.e., 400 and 150 Veh/h). Te left-turn volume varied in the range of 100–400 Veh/h with an incremental step of 50 Veh/h. Fig- 4.4. Dual Left-Turn Lanes. Te through and right-turn ure 6 shows that the intersection delay increased as the left- volumes were considered constant for all directions (i.e., 400 turn volume grew. Tis is true for both simulation and and 150 Veh/h). Te left-turn volume varied between 100 HCM2016 results. Comparing the results of simulation and 700 Veh/h with an incremental step of 50 Veh/h. Fig- software and HCM2016 showed that the intersection delay ure 8 shows that the intersection delay increased as the left- diference was neglectable for the trafc volumes of turn volume grew. Tis is true for both simulation and 100–350 Veh/h. However, the diference became 10% and HCM2016 results. Comparing simulation and HCM2016 34.6% for the volumes of 350 and 400 Veh/h, respectively. results indicated that the intersection delay diference was Tus, the volume of 350 Veh/h was the threshold for a 17% for the trafc volumes of 100–400 Veh/h. However, the sudden change in delay. diference decreased at the range of 450–650 Veh/h, and for the volumes of 550, 600, and 650 Veh/h, the diference became insignifcant. For the trafc volume of 700 Veh/h, 4.3. Dual Right-Turn Lanes. Figure 7 shows the efect of the the diference between the simulation results and HCM2016 number of right-turn lanes on the dual-lane intersection reached 11.4%. delay. Te right-turn volume varied between 200 and 1,450 Veh/h with an incremental step of 50 Veh/h. Te simulation results showed that the delay parameter initially 4.5.ComparisonofSingleandDualLanes. Figure 9 shows the decreased for the trafc volume up to 550 Veh/h. For higher volumes, the delay parameter increased. Tere was a 39.8% efect of the number of right-turn lanes on the intersection diference between the simulation results and HCM2016 at delay estimated by the simulation and HCM2016. As the the volume of 550 Veh/h. HCM2016 results indicated that number of right-turn lanes increased to two, the HCM2016 the delay was independent of the incoming volume below did not estimate any changes in the intersection delay for the 950 Veh/h. Te delay increased as the volume increased from single right-turn lane with a volume below 550 Veh/h and 950 to 1,450 Veh/h. For the trafc volume of 1,200 Veh/h, dual right-turn lanes with a volume below 1,000 Veh/h. On the delay reported by the simulation and HCM2016 was the the other hand, the diference between the delays reported by same. As the volume increased from 1,200 to 1,450 Veh/h, HCM2016 was neglectable for single and dual right-turn lanes at trafc volumes below 500 Veh/h. However, the the growing trend of the delay in the simulation was more than that of HCM2016. For instance, the delay parameter simulation results indicated that the delay was reduced for Delay (s) 8 Journal of Advanced Transportation 50 100 150 200 250 300 350 400 450 Volume (Veh/h) simulation HCM 2016 Figure 6: Te efect of volume variations on delay for single left-turn lane. 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400 1500 Volume (Veh/h) simulation HCM 2016 Figure 7: Te efect of volume variations on delay for dual right-turn lane. 800 Veh/h. At 1,450 Veh/h, the delay was 23 s for the dual Table 4: F-test performed between simulation and HCM2016. right-turn lanes. Te trafc scenarios of signalized intersections were Methods Variance F P value F critical analyzed for diferent V/C values with a maximum of 1. Simulation 19.44 6.04 0.001 1.95 Figure 10 shows the simulation results of the single and dual HCM2016 3.21 right-turn lanes. As the volume increased from 100 to 300 Veh/h, V/C remained fxed at 0.39 for the single right- turn lane. For trafc volumes in the range of 300–800 Veh/h, dual right-turn lanes with volumes in the range of V/C suddenly increased to 0.89. For the dual right-turn 100–550 Veh/h. For higher volumes up to 1,450 Veh/h, the lanes, V/C was constantly equal to 0.39 for the trafc volume delay increased. Tis was the same for the single right-turn of 100 to 600 Veh/h. For higher volumes, V/C gradually lane. Te only diference was that the delay parameter de- increased and reached 0.91 at 1,450 Veh/h. Comparing the creased in the range of 100–300 Veh/h and then increased single and dual right-turn lanes suggested that, similar to the for higher volumes. Te simulation results showed that in delay parameter, V/C converged faster to 1 at low trafc the range of 100–300 Veh/h, the delay associated with the volumes when using a single lane, but for large trafc single and dual lanes was equal. A signifcant diference was volumes, the convergence speed was higher when using dual observed for volumes higher than 300 Veh/h. For a single lanes. Tis is due to the efect of the number of right-turn right-turn lane, the delay was 23 s for the trafc volume of Delay (s) Delay (s) Journal of Advanced Transportation 9 100 200 300 400 500 600 700 Volume (Veh/h) simulation HCM 2016 Figure 8: Te efect of volume variations on delay for dual left-turn lane. 50 250 450 650 850 1050 1250 1450 1650 Volume (Veh/h) Simulation (2-lane) HCM2016 (2-lane) Simulation (1-lane) HCM2016 (1-lane) Figure 9: Delay variations of the efect of changing trafc scenarios in single and dual right-turn lanes. 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 50 150 250 350 450 550 650 750 850 950 1050 1150 1250 1350 1450 Volume (Veh/h) V/C Right-turn 1-lane V/C Right-turn2-lane Figure 10: Comparison of changing a single and dual right-turn lane in volume to capacity (v/c) ratio. Delay (s) Max (V/C) Delay (s) 10 Journal of Advanced Transportation 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 Volume (Veh/h) simulation (1 lane) HCM 2016 (1 lane) simulation (2 lane) HCM 2016 (2 lane) Figure 11: Delay variations of the efect of changing trafc scenarios in a single and dual left-turn lane. 1.2 0.8 0.6 0.4 0.2 0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 Volume (Veh/h) Lef-turn 1-lane Lef-turn 2-lane Figure 12: Comparison of changing single to dual left-turn lane on volume to capacity (v/c) ratio. lanes. For example, for V/C � 0.89, the volume was 800 Veh/ Figure 12 depicts the simulation results of the single and h for a single lane and 1,400 Veh/h for dual lanes (i.e., 75% dual left-turn lanes. As the left-turn volume increased from higher trafc volume). 50 to 100 Veh/h, V/C remained constant at 0.31 for the single Figure 11 illustrates that by increasing the number of left- right-turn lane. For trafc volumes in the range of turn lanes to two, the delay was 20 s for a single left-turn lane at 100–400 Veh/h, V/C increased to 0.98. For the dual right- 350 Veh/h, while the same delay was obtained by the dual left- turn lanes, V/C was constantly equal to 0.28 for the trafc turn lanes at 700 Veh/h, according to HCM2016. Te simulation volume of 50 to 200 Veh/h. For higher volumes, V/C results indicated that increasing the number of left-turn lanes led gradually increased and reached 0.89 at 700 Veh/h. Com- to a delay reduction from 45.6 to 10.3 s for the left-turn volume paring the single and dual left-turn lanes suggested that, of 400 Veh/h. Tis shows that even for smaller volumes (around similar to the delay parameter, V/C converged faster to 1 at 400 Veh/h) in which the signalized intersection was unsaturated, low trafc volumes when using a single lane, but for large there was a signifcant diference in delay parameter between the trafc volumes, the convergence speed was higher when single and dual left-turn lanes in both simulation results and using dual lanes. For example, for V/C � 0.86, the volume HCM, respectively, 77.4% and 59.7%. Terefore, the diference was 350 Veh/h for a single lane and 700 Veh/h for dual lanes can be attributed to the number of lanes. (i.e., 100% higher trafc volume). Max (V/C) Delay (s) Journal of Advanced Transportation 11 depending on the number of right-turn lanes and Table 5: t-test performed between the single and dual lanes. volumes. Furthermore, at low volumes, the delay Turn type Method t stat P value t Critical value decreased and then increased in both single Right- and dual right-turn lanes. Simulation and HCM2016 −3.96 0.001 2.02 turn Left-turn Simulation and HCM2016 −2.21 0.03 2.1 Overall, if the intersection-occupied area is enough, increasing the number of right- and left-turn lanes can be a suitable alternative for at-grade signalized intersections to Te diference between the single and dual lanes (both improve the throughput of an intersection. Our results can right and left turn) results was verifed statistically using the ofer a good vision for trafc engineers to modify the ge- two-sample t-test method. Table 5 shows a signifcant dif- ometry of the four-leg undersaturated signalized intersec- ference between single and dual lanes in volume results at a tions under conditions that the right or left-turn volumes are 95% confdence level (P value <0.05). incremental, and the through volumes are constant. AASHTO 2018 and NCHRP 780 provide recommen- dations based on driving behavior and trafc parameters for Data Availability the length of right- and left-turn lanes in a given condition of geometric elements of at-grade signalized intersections. No datasets were generated or analyzed during the current Also, previous studies conducted in this feld have examined study. the length of the lane in diferent turning movements of an approach (see Table 1). In contrast, the result analysis of this research shows that the number of trafc lanes can signif- Conflicts of Interest icantly impact increasing the throughput of intersections. In Te authors declare that they have no conficts of interest. this study, we have focused on the efect of the number of turning lanes on the volume changes of right turn and left turn. Analyses in undersaturated conditions (V/C< 1) References showed that the HCM2016 delay model is not sensitive to [1] M. E. M. A. Al-Omari and M. Abdel-Aty, “Evaluation of a volume changes, especially in right-turning movements, up New Intersection Design,“Shifting Movements,” Trans- to a specifc value. While much of the research reviewed in portation Research Record, vol. 2675, 2021. the background has been done for V/C � 1 or V/C> 1, this [2] S. Kikuchi, M. Kii, and P. Chakroborty, “Lengths of double or insensitivity of HCM is not seen in saturated and over- dual left-turn lanes,” Transportation Research Record, saturated conditions. vol. 1881, no. 1, pp. 72–78, 2004. [3] M. Liu, L. Xu, L. Shen, and S. Jin, “Modeling capacity at 5. Conclusion signalized intersections with a left-turn storage bay consid- ering signal timing plan,” Journal of Transportation Engi- Tis study sought to assess the performance of signalized neering, Part A: Systems, vol. 145, no. 2, Article ID 04018084, intersections by changing the number of right- and left- 2019. turn approaches and trafc scenarios for a symmetrical [4] S. L. Hallmark and N. Hawkins, “Adding Turn Lanes/ Channelization Tech Brief,” 2014, https://www.intrans.iastate. four-leg intersection using Synchro. For all trafc sce- edu/. narios, the incoming volume of the through approach [5] Y. Guo, P. Liu, Y. Wu, and J. Chen, “Evaluating how right- was considered constant, and the channelized single or turn treatments afect right-turn-on-red conficts at signalized dual right-turn lane approaches and the left-turn volume intersections,” Journal of Transportation Safety and Security, approach with single or dual lane storage length were vol. 12, no. 3, pp. 419–440, 2020. increased under unsaturated trafc conditions. Te fol- [6] E. Nyame-Baaf, C. A. Adams, and K. K. Osei, “Volume lowing conclusions can be drawn: warrants for major and minor roads left-turning trafc lanes at unsignalized T-intersections: a case study using VISSIM (i) As the number of left-turn lanes increased, the re- modelling,” Journal of Trafc and Transportation Engineering, sults showed that for low trafc volumes (less than vol. 5, no. 5, pp. 417–428, 2018. 400 Veh/h) in which the signalized intersection was [7] K. Fitzpatrick, W. H. Schneider, and E. S. Park, “Operation unsaturated, the diference in delay parameter be- and safety of right-turn lane designs,”TransportationResearch tween the single and dual left-turn lanes was sig- Record, vol. 1961, no. 1, pp. 55–64, 2006. nifcant in simulation and HCM2016, respectively, [8] W. Guo, W. Wang, G. Wets et al., “Infuence of stretching- 77.4% and 59.7%. segment storage length on urban trafc fow in signalized intersection,” International Journal of Computational Intel- (ii) Generally, as the through approach volume ligence Systems, vol. 4, no. 6, pp. 1401–1406, 2011. remained constant, the delay parameter reported by [9] S. Farivar, R. Andalibian, and Z. Z. Tian, “Modeling right-turn HCM2016 was fxed up to a certain threshold (10.8 s) blockage and approach capacity at signalized intersections for increased channelized right-turn volume. Te with channelized right-turn lanes,” Journal of Transportation delay parameter of the undersaturated intersection Engineering, Part A: Systems, vol. 143, no. 2, Article ID was less sensitive to single and dual right-turn lane 04016006, 2017. changes. However, the simulation results indicated [10] F. Dion, H. Rakha, and Y. S. Kang, “Comparison of delay that the delay parameter was not constant, estimates at under-saturated and over-saturated pre-timed 12 Journal of Advanced Transportation signalized intersections,” Transportation Research Part B: [27] R. Yao, “Settings of short left-turn lane and signal phase Methodological, vol. 38, no. 2, pp. 99–122, 2004. sequence for isolated signalized intersections,” Transport, vol. 31, no. 4, pp. 416–426, 2014. [11] R. Yao and H. Michael Zhang, “Optimal allocation of lane [28] W. Ma, Y. Liu, J. Zhao, and N. Wu, “Increasing the capacity of space and green splits of isolated signalized intersections with signalized intersections with left-turn waiting areas,” Trans- short left-turn lanes,” Journal of Transportation Engineering, portation Research Part A: Policy and Practice, vol. 105, vol. 139, no. 7, pp. 667–677, 2013. pp. 181–196, 2017. [12] HCM, Highway Capacity Manual. Transportation Research [29] S. Kikuchi and N. Kronprasert, “Determining the length of the Board, National Research Council, Washington, D.C., USA, right-turn lane at a signalized intersection,” Transportation Research Record, vol. 2060, no. 1, pp. 19–28, 2008. [13] AASHTO (American Association of State Highway and [30] G. S. Macfarlane, M. Saito, and G. G. Schultz, “Delay under- Transportation ofcials), A Policy on Geometric Design of estimation at free right-turn channelized intersections,” Pro- Highways and Streets, AASHTO, Washington”, D.C. USA, cedia-social and behavioral sciences, vol. 16, pp. 560–567, 2011. [31] X. Chen, Y. Qi, and D. Li, “Estimating right-turn-on-red capacity [14] F. Wang, K. Li, C. Shao, J. Zhang, B. Li, and N. Han, “Op- for dual right-turn lanes at signalized intersections,” Trans- timisation of signal timing at intersections with waiting portation Research Record, vol. 2286, no. 1, pp. 29–38, 2012. areas,” Promet - Trafc and Transportation, vol. 34, no. 2, [32] X. Chen, Y. Qi, and G. Liu, “Empirical study of gap-accep- pp. 195–209, 2022. tance behavior of right-turn-on-red drivers on dual right-turn [15] R. Guo, J. Liu, and Y. Qi, “An innovative signal timing strategy lanes,” Journal of Transportation Engineering, vol. 139, no. 2, for implementing contrafow left-turn lanes at signalized pp. 173–180, 2013. intersections with split phasing,”Sustainability, vol. 13, no. 11, [33] B. Li, W. Cheng, Y. Bie, and B. Sun, “Capacity of advance p. 6307, 2021. right-turn motorized vehicles at signalized intersections for [16] J. Wu, P. Liu, Z. Z. Tian, and C. Xu, “Operational analysis of mixed trafc conditions,” Mathematical Problems in Engi- the contrafow left-turn lane design at signalized intersections neering, vol. 2019, Article ID 3854604, 13 pages, 2019. in China,” Transportation Research Part C: Emerging Tech- [34] S. Kikuchi, N. Kronprasert, and M. Kii, “Lengths of turn lanes nologies, vol. 69, pp. 228–241, 2016. on intersection approaches: three-branch fork lanes–left-turn, [17] Z. Li and L. Elefteriadou, “Maximizing the trafc throughput through, and right-turn lanes,” Transportation Research Re- cord, vol. 2023, no. 1, pp. 92–101, 2007. of turn bays at a signalized intersection approach,” Journal of [35] D. Zupanovi´ c, M. Anzek, ˇ and G. Kos, “Optimisation of signal- Transportation Engineering, vol. 139, no. 5, pp. 425–432, 2013. controlled intersection capacity,” Promet - Trafc and [18] Y. Zheng, X. Hua, W. Wang, J. Xiao, and D. Li, “Analysis of a Transportation, vol. 22, no. 6, pp. 419–431, 2012. signalized intersection with dynamic use of the left-turn lane [36] W. Wu, W. Ma, and K. Long, “Capacity matching based for opposite through trafc,” Sustainability, vol. 12, no. 18, model for protected left turn phases design of adjacent sig- p. 7530, 2020. nalized intersections along arterial roads,” Promet - Trafc [19] I. Shatnawi, P. Yi, and I. Khliefat, “Automated intersection and Transportation, vol. 27, no. 1, pp. 13–21, 2015. delay estimation using the input–output principle and turning [37] H. K. Kim, S. K. Oh, and G. M. Bae, “Demonstrative com- movement data,” International Journal of Transportation parison of two trafc signal optimization programs: synchro-6 Science and Technology, vol. 7, no. 2, pp. 137–150, 2018. and PASSER II,” KSCE Journal of Civil Engineering, vol. 10, [20] A. Abdi Kordani, O. Rahmani, A. S. Abdollahzadeh Nasiri, no. 1, pp. 41–45, 2006. and S. M. Boroomandrad, “Efect of adverse weather con- [38] H. Xu, H. Liu, and Z. Tian, “Control delay at signalized diamond ditions on vehicle braking distance of highways,” Civil En- interchanges considering internal queue spillback,” Trans- gineering Journal, vol. 4, no. 1, pp. 46–57, 2018. portation Research Record, vol. 2173, no. 1, pp. 123–132, 2010. [21] Y. Qi, L. Yu, M. Azimi, and L. Guo, “Determination of storage [39] C. Zheng, P. Liu, J. J. Lu, and H. Chen, “Evaluating the efect lengths of left-turn lanes at signalized intersections,” Trans- efects of U-turns on level of service of signalized intersections portation Research Record, vol. 2023, no. 1, pp. 102–111, 2007. using synchro and SimTrafc,” in Proceedingsofthe2009IEEE [22] K. Yin, Y. Zhang, and B. X. Wang, “Analytical models for Intelligent Vehicles Symposium, pp. 971–976, IEEE, Xian, protected plus permitted left-turn capacity at signalized in- China, June 2009. tersection with heavy trafc,” Transportation Research Record, [40] J. Zheng, H. X. Liu, S. Misgen, K. Schwartz, B. Green, and vol. 2192, no. 1, pp. 177–184, 2010. M. Anderson, “Use of event-based trafc data in generating [23] S. Kikuchi and N. Kronprasert, “Determining lengths of left- time–space diagrams for evaluation of signal coordination,” turn lanes at signalized intersections under diferent left-turn TransportationResearchRecord, vol. 2439, no. 1, pp. 94–104, 2014. signal schemes,” Transportation Research Record, vol. 2195, [41] H. Hatami and I. Aghayan, “Trafc efciency evaluation of elliptical roundabout compared with modern and turbo no. 1, pp. 70–81, 2010. roundabouts considering trafc signal control,” Promet - [24] W. L. Reynolds, X. Zhou, N. M. Rouphail, and M. Li, “Es- Trafc and Transportation, vol. 29, no. 1, pp. 1–11, 2017. timating sustained service rates at signalized intersections [42] A. R. Dabiri, I. Aghayan, and F. Hadadi, “A comparative with short left-turn pockets: mesoscopic approach,” Trans- analysis of the performance of turbo roundabouts based on portation Research Record, vol. 2173, no. 1, pp. 64–71, 2010. geometric characteristics and trafc scenarios,” Trans- [25] K. Yin, Y. Zhang, and B. X. Wang, “Modeling delay during portation letters, vol. 13, no. 9, pp. 674–685, 2021. heavy trafc for signalized intersections with short left-turn [43] K. Fitzpatrick, M. A. Brewer, P. Dorothy, and E. S. Park, bays,” Transportation Research Record, vol. 2257, no. 1, “Design Guidance For Intersection Auxiliary Lanes (No. pp. 103–110, 2011. Project 03-102,” NCHRP 780, 2014, https://www.trb.org/ [26] Y. G. Qi, L. Guo, L. Yu, and H. Teng, “Estimation of design Publications/Blurbs/171328.aspx., 2014. lengths of left-turn lanes,” Journal of Transportation Engi- neering, vol. 138, no. 3, pp. 274–283, 2012. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Advanced Transportation Hindawi Publishing Corporation

The Effect of the Number of Right-Turn and Left-Turn Lanes on the Performance of Undersaturated Signalized Intersections

Download PDF
12 pages

Loading next page...
 
/lp/hindawi-publishing-corporation/the-effect-of-the-number-of-right-turn-and-left-turn-lanes-on-the-wy5uaGVTnV

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Hindawi Publishing Corporation
ISSN
0197-6729
eISSN
2042-3195
DOI
10.1155/2023/8764498
Publisher site
See Article on Publisher Site

Abstract

Hindawi Journal of Advanced Transportation Volume 2023, Article ID 8764498, 12 pages https://doi.org/10.1155/2023/8764498 Research Article TheEffectoftheNumberofRight-TurnandLeft-TurnLanesonthe Performance of Undersaturated Signalized Intersections 1 2 1 Omid Rahmani , Amir Saman Abdollahzadeh Nasiri , and Iman Aghayan Faculty of Civil Engineering, Shahrood University of Technology, Shahrood, Iran Department of Civil Engineering, South Tehran Branch, Islamic Azad University, Tehran, Iran Correspondence should be addressed to Omid Rahmani; omid.rahmani@shahroodut.ac.ir Received 6 August 2022; Revised 17 October 2022; Accepted 23 January 2023; Published 16 February 2023 Academic Editor: David Rey Copyright © 2023 Omid Rahmani et al. Tis is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Vehicle trafc fow channelizing can signifcantly contribute to fewer collisions and smooth trafc fow in isolated signalized intersection design. Terefore, this study investigates the efect of changes in the number of right-turn lanes with splitter islands and left-turn storage bays for diferent volume approaches on intersection performance. Tis study aims to provide an accurate analysis to estimate the efect of such changes on signalized intersections, which can improve intersection performance. Furthermore, the delay parameter derived from HCM2016 and simulation was evaluated for diferent scenarios. For this purpose, simulation was performed using Synchro software. In this study, a symmetrical, 90-degree, undersaturated (volume per capacity lower than 1, V/C < 1), and four-leg intersection were considered according to AASHTO Green Book 2018 suggestions. HCM2016 results indicated the delay parameter was less sensitive to the undersaturated condition than the right- turn volume variation for both single and dual-lane bays. However, the simulation results indicated that the delay parameter was not constant, depending on the number of right-turn lanes and volumes. On the other hand, there was a signifcant diference in delay parameters between the single and dual left-turn lanes for both simulation and HCM cases estimated 77.4% and 59.7%, respectively. Overall, this study can provide a vision for trafc engineers to modify the geometry of the four-leg signalized intersections, if the right-turn or left-turn demand volume of an undersaturated signalized intersection is larger than the through approach. of the intersection is considered when it is not feasible to 1. Introduction reduce the delay time of a heavy trafc intersection through Te intersection capacity plays an important role in de- trafc light timing and phasing. Right-turn splitter islands termining the trafc capacity of an urban road network. and left-turn storage bays can be alternative geometric Regarding their heavy trafc, signalized intersections have characteristics to improve the signalized intersections. Tese been the focus of many studies because a large portion of a techniques can signifcantly reduce collisions and contribute to smooth trafc fow [4, 5]. vehicle’s lost time is related to signalized intersections. Te main methods to reduce delays and improve the intersection Trafc fow has been less considered as a statistical capacity are to optimize the trafc light timing and modify criterion to change left- and right-turn volumes [6]. If ac- the geometric characteristics of the intersections [1]. curate predictions can be made to estimate the efect of Safety and capacity play the most important role in the geometric modifcations on left and right turns, a better geometric design of an isolated signalized intersection [2, 3]. vision can be outlined for the design of signalized inter- Trafc signal timing and phasing optimization without sections before saturation. Te geometric parameters for the changing the geometric characteristics of the intersection are design of the left and right turns in a signalized intersection possible to limited values. Terefore, geometric modifcation include bay width, channelized right-turn radius, right-turn 2 Journal of Advanced Transportation lane length, left-turn storage length, taper length, and the undersaturated isolated signalized intersection considering number of lanes [7]. Most studies have considered right-turn the geometric characteristics (i.e., number of lanes) of channelized right-turn approaches and the storage length of lane length and left-turn storage length, while the increased number of lanes and considering the storage or deceleration the left-turn approach. length can prevent lane blockage for incremental turning A summary of some previous studies related to signal- volume of the intersections [8, 9]. ized intersections is shown in Table 1. Based on the Te majority of studies have been dedicated to over- movement in diferent signalized intersection approaches, saturated (i.e., V/C≥ 1) or saturated signalized intersections the intersection assessment method (probability, simulation, [10]. Signalized intersection phasing and timing can be a or both), trafc fow parameters (queue length, delay, and V/ complex task in some cases due to the saturated incoming C), geometric parameters, and also channelized right-turn intersection trafc [11]. It is necessary to avoid saturation as movement were reviewed. Te HCM-based studies are also much as possible while designing and selecting the geo- presented. Tese studies propose a model or process to metric alternatives of an intersection. complement the HCM methodology or compensate for its weaknesses in signalized intersections. Te literature shown According to the Highway Capacity Manual (HCM) 2016 regulation, the delay is the main criterion (measure of in Table 1 is categorized based on the movement, including efectiveness (MOE)) for at-grade intersection efciency (1) left-turn lane, (2) right-turn lane, (3) right turn, left turn, evaluation [12]. Based on HCM2016, the delay calculation and through, and (4) right turn, left turn, through, and methodology for undersaturated signalized intersections is U-turn. less sensitive to changing the number of channelized right- Table 1 indicates that the geometric parameter evalua- turn lanes. In other words, the delay remained constant for tions aimed to improve the capacity of signalized inter- right-turn overdemand and fxed through approach volume. sections were focused on left- and right-turn storage bays. Also, the extent to which the changing number of bays Te simultaneous efect of geometric and trafc parameter afects the intersection delay is unknown. variations, including the number of right- and left-turn lanes Terefore, this study seeks to evaluate the efect of the and their trafc demand volume, was not included. Tese studies were carried out for near-saturated or oversaturated right-turn lane and left-turn storage bay quantity on the capacity of an undersaturated signalized intersection. To this signalized intersections. If the left- and right-turn volume of end, a symmetrical 90-degree four-leg signalized intersec- a signalized intersection increases, it is important to rec- tion without longitudinal grade was designed based on the ognize its limitations and maintain the intersection in an AASHTO guideline [13]. Te efect of the number of lanes undersaturated state. on the delay parameter was compared to the HCM2016 Te frst novelty of this study is to evaluate the efect of model. Te trafc and geometric characteristics of the un- the number of left- and right-turn lanes with storage length dersaturated isolated signalized intersection were modeled on the undersaturated signalized intersection capacity. Te and analyzed by Synchro11 software. Te analyses were assessment was performed for trafc volume scenarios of based on the delay and volume per capacity (V/C) ratio. variant left- and right-turn approaches and constant through approach volume. Tus, the combined efect of the number of lanes and trafc fow (left- and right-turn volume) was 2. Literature Review investigated. HCM assumes a constant delay for the un- Previous studies have focused on evaluating the perfor- saturated state up to a certain right-turn volume. Terefore, mance of signalized intersections with left- and right-turn another novelty of the study is to evaluate the delay and V/C storage bays and channelized right-turn lanes in terms of variations as the main criteria to determine the signalized delay, V/C, and queue length. In general, the research re- intersection capacity relative to HCM. garding the capacity of signalized intersections with storage bays can be divided into two categories: (i) the phasing and 3. Methodology timing of signalized intersections and (ii) the efect of geometric characteristics on the capacity of signalized Synchro trafc simulator was employed to defne the geo- intersections. metric characteristics, trafc volumes, trafc light timing, For the frst category, it was assumed that the capacity of and delay. Many studies regarding trafc in at-grade sig- an isolated signalized intersection was improved by phasing nalized intersections used Synchro [35–40]. By considering and timing optimization [11, 14–16]. In the second category, diferent HCM criteria [41, 42], Synchro was selected be- the intersection throughput was improved by changing the cause it contains the sixth edition of HCM (HCM2016). Te geometric characteristics of the left and right turns, such as average control delay (d) for each lane is based on the right-turn channelization and increased left- or right-turn following formula: storage length [17–20]. Te current study investigates an 􏽳���������������� � 3600 (3600/c)x ⎡ ⎢ ⎤ ⎥ ⎢ ⎥ ⎣ ⎦ (1) d � + 900 T (x − 1) + (x − 1) + + 5 × min[x, 1], c 450 T Journal of Advanced Transportation 3 Table 1: Studies related to signalized intersections. Evaluation methods Trafc parameters Movement HCM Proposed Geometric Right-turn Study (year) type checked model parameters channelized Simulation Probability Queue Delay v/c Kikuchi et al. Length of left- — — ✓ ✓ — — — — (2004) [2] turn lane Qi et al. (2007) Length of left- — — ✓ ✓ — — ✓ — [21] turn lane Yin et al. (2010) Length of left- ✓ CORSIM ✓ ✓ — — ✓ — [22] turn lane Kikuchi and Monte Carlo Length of left- Kronprasert Left-turn lane — method and ✓ ✓ — — ✓ — turn lane (2010) [23] VISSIM Reynolds et al. VISSIM Length of left- ✓ — ✓ — ✓ ✓ — (2010) [24] DYNASMART turn lane Guo et al. Length of left- — — ✓ — — ✓ ✓ — (2011) [8] turn lane Yin et al. (2011) Length of left- — VISSIM ✓ ✓ ✓ — ✓ — [25] turn lane HSC+ Qi et al. (2012) Synchro Length of left- ✓ — ✓ — — — — [26] VISSIM turn lane SimTrafc Yao and Length of left- Michael Zhang ✓ VISSIM — ✓ ✓ — ✓ — turn lane (2013) [11] Length of left- Yao (2016) [27] ✓ — ✓ — ✓ ✓ ✓ — turn lane Number of Ma et al. (2017) left-turn lane ✓ VISSIM ✓ — — ✓ ✓ — [28] in waiting area Liu et al. (2018) Length of left- ✓ VISSIM ✓ — — ✓ ✓ — [3] turn lane Zheng et al. Length of left- ✓ VISSIM — ✓ ✓ — ✓ — (2020) [18] turn lane Guo et al. Length of left- — VISSIM — ✓ — — — — (2021) [15] turn lane Kikuchi & Length of Kronprasert — — ✓ ✓ — — — right-turn — (2008) [29] lane Macfarlane Synchro et al. (2011) ✓ — — ✓ — ✓ — — (SimTrafc) [30] Chen et al. ✓ VISSIM ✓ — ✓ ✓ ✓ — — (2012) [31] Chen et al. Right-turn ✓ — ✓ — — — ✓ — ✓ (2013) [32] lane Length of Farivar et al. ✓ VISSIM ✓ — ✓ — ✓ right-turn ✓ (2016) [9] lane Length of Farivar et al. ✓ VISSIM ✓ — — ✓ ✓ right-turn ✓ (2017) [9] lane Li et al. (2019) ✓ VISSIM — — — ✓ ✓ — ✓ [33] Length of Kikuchi et al. ✓ VISSIM ✓ ✓ — — — right-turn and ✓ (2007) [34] left-turn lane Right turn, Li and left turn, and Elefteriadou ✓ CORSIM — ✓ ✓ — — — — through (2013) [17] Shatnawi et al. ✓ VISSIM — ✓ ✓ — — — — (2018) [19] 4 Journal of Advanced Transportation Intersection geometry Traffic demand Signalized intersection Intersection angle = 90° Traffic volumes: (Without longitudinal grades) Through= 400 (Veh/h) Single Right-turn lane = Initial signal cycle length 100 to 800 (Veh/h) Storage length for right-turn Signal phasing arrangement Dual Right-Turn Lanes = and left-turn (Based on (two-phases) 200 to 1450 (Veh/h) AASHTO 2018) Single Left-turn Lane = 100 to 400 (Veh/h) Single and dual lane for right Dual Left-turn Lane = 100 and left-turn to 700 (Veh/h) Right-turn curb radius (15.3 m) Synchro (Simtraffic) Lane width (3.65 m) Optimize signal cycle Critical gap = 5.0 (s) and length for each scenario Follow-up gap = 2.2 (s) (Based on AASHTO 2018) Simulation HCM 2016 Comparing outputs analysis outputs between Control (V/C) <1 single and dual- Delay (s) lane Delay (s) Control all demand volume (V/C) change scenarios for under saturation Comparing outputs between simulation and HCM Figure 1: Research method in detail. where x � volume-to-capacity ratio of each lane, c � capacity Figure 1 summarizes the defned inputs and outputs for of the subject lane (veh/h), and T � time period (h) � 0.25 simulation and the methodology details. Te drivers’ be- (h). havior in terms of gaps was selected based on AASHTO 2018 Te control delay for an approach is calculated using the selections. To this end, critical gaps and follow-up gaps were weighted average of the delay for each lane on the approach, accounted for in the simulations. Te results were used to weighted by the volume in each lane. Te calculation is compare the delay obtained by the simulation and HCM. demonstrated in Te volume per capacity (V/C) obtained by the simulation was evaluated for each scenario to control whether the in- 􏽐 d v i i d � , (2) coming right- and left-turn volumes were in an unsaturated 􏽐 v state (V/C< 1). A single and dual-lane method was adopted to compare the right- and left-turn movements. where d � control delay for the approach (s/veh), d � a i control delay for lane i (s/veh), and v � fow rate for lane i (veh/h). 3.1. Geometric Design of Intersections. For all designed Te control delay for the intersection as a whole is intersections, the intersections were 90 degrees without a similarly calculated by computing a weighted average of the longitudinal slope. Te radius of the channelized right- delay for each approach and weighted by the volume on each turn horizontal arc edge was 15.3 m. Te ideal lane width approach which is shown in the following equation: was 3.65 m. Te length of the left-turn storage bay was 􏽐 d v based on NCHRP 780 [43], which is suggested by the a a d � , (3) intersection latest version of the AASHTO Green Book (2018). To this 􏽐 v end, the left-turn storage bay was considered 30.5 m long where d � control delay for the entire intersection intersection because the through movement volume of the opposite (s/veh), d � control delay for approach a (s/veh), and v � a a trafc fow was 400 Veh/h. However, since 2% of the total fow rate for approach a (veh/h). incoming trafc was associated with heavy vehicles, 7.6 m Journal of Advanced Transportation 5 Right-Turn Channelized Island (Splitter Island) Storage Length Right-Turn Lane Storage (deceleration) Curb Radius Length Left-Turn Lane (Reserved Lane) Length of Taper (a) (b) Figure 2: Geometric design elements for turning movement at an intersection: (a) single right-turn lane and (b) single left-turn lane. was added to the left-turn storage bay length, according 4. Results’ Analysis and Discussion to AASHTO 2018, leading to a 38.1 m long left-turn After the scenario simulation, the results were analyzed and storage bay. compared. Te results were estimated for a single right-turn Figures 2(a) and 2(b) show the length of the right-turn lane, a single left-turn lane, a dual right-turn lane, and a dual bays and left-turn storage length after the geometric mod- left-turn lane. In addition, the results were presented in the ifcations. All the scenarios were considered in separate form of volume-delay control and V/C curves to analyze the phasing. First, the trafc light optimization was carried out intersection performance and compare diferent scenarios. by optimizing the intersection cycle length, green light Te scenarios were simulated for unsaturated trafc using timing for each route, and approaches. Te optimization was Synchro, and the results were compared to HCM2016. Tus, performed by the Synchro simulator. Ten, the total delay of the following results were obtained. the intersection was evaluated. Te undersaturated state (V/ C< 1) was the criterion for selecting the defned light or heavy trafc volumes. In addition to one right- and left-turn lane, dual right- and left-turn lanes were also addressed in 4.1. Single Right-Turn Lane. Te through and left-turn the current study. volumes were assumed constant for all directions (i.e., 400 and 150 Veh/h). Te right-turn volume varied in the range of 100–800 Veh/h with an incremental step of 50 Veh/h. Te 3.2. Trafc Demand Characteristics at the Intersection. simulation results showed that the delay parameter initially Two general volumetric cases with V/C< 1 were considered decreased for the trafc volume up to 300 Veh/h (Figure 5). to defne the scenarios. In the frst case, the through and left- For higher volumes, the delay parameter increased. turn approach volumes were fxed (400 and 150 Veh/h, HCM2016 result indicated that the delay was independent of respectively). Te volume growth rate was defned as 50 Veh/ the incoming volume below 500 Veh/h. As the volume in- h for each new scenario for right-turn movements. Te creased from 500 to 600 Veh/h, the delay increased, and for single right-turn and dual right-turn lane volumes were in higher volumes, the amount of delay increased signifcantly. the range of 100–800 and 200–1, 450 Veh/h, respectively. In For the trafc volume of 600 Veh/h, the delay reported by the second case, the through and right-turn approach vol- our simulations and HCM2016 was equal. For over 600 umes were fxed (400 and 150 Veh/h, respectively). Te (Veh/h), the delay of both HCM and simulation increases volume growth rate was defned as 50 Veh/h for each new because the delay of the single right-turn lane is sensitive to scenario for left-turn movements. Te single left-turn lane the incoming volume increase. In fact, the volume of 600 Veh/ and dual left-turn lane volumes were in the range of 50–400 h was the threshold for a sudden change in delay. As the and 200–1, 450 Veh/h, respectively (Table 2). After applying volume increased from 600 to 800 Veh/h, the growing trend the passing volume, the trafc light cycle length was opti- of the delay in the simulation was more drastic than that of mized by Synchro. Figure 3 shows the intersection plans for HCM2016, indicating that HCM2016 is a more conservative dual left-turn lanes (3a), dual right-turn lanes (3b), and method for delay estimation. For instance, the delay pa- single left- and right-turn lanes (3c). Figure 4 depicts the 3D rameter obtained by simulations was 11% higher at 650 Veh/h, view of the simulation of a scenario. while the diference was 26.7% at the volume of 800 Veh/h. 6 Journal of Advanced Transportation 200 200 400 400 150 150 (a) (b) (c) Figure 3: Sample of simulated scenarios in Synchro: (a) dual left-turn lanes, (b) dual right-turn lanes, and (c) single right- and left-turn lanes. Figure 4: Simulation of intersection in Synchro (3D). Table 2: Summary of scenarios, trafc demand volume, and outputs. Type of Number of Trough trafc Left-turn trafc volume Right-turn trafc volume (Veh/h) Outputs turn lanes volume (Veh/h) (Veh/h) 100, 150, 200, 250, 300, 350, 400, 450, 500, 550, Single lane 400 150 600, 650, 700, 750, 800 Right- 200, 250, 300, 350, 400, 450, 500, 550, 600, 650, turn Dual lane 400 700, 750, 800, 850, 900, 950, 1000, 1050, 1100, 150 Delay and 1150, 1200, 1250, 1300, 1350, 1400, 1450 volume to 50, 100, 150, 200, 250, Single lane 400 150 capacity (V/C) 300, 350, 400 Left- 100, 150, 200, 250, 300, turn Dual lane 400 150 350, 400, 450, 500, 550, 600, 650, 700 In addition, the diference between the simulation and lane. According to Table 3, the results demonstrate that there is HCM2016 control delay results was verifed statistically using a a signifcant diference between simulation and HCM2016 in two-sample F-test for variances method in a single right-turn control delay results at a 95% confdence level (P value <0.05). 150 Journal of Advanced Transportation 7 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800 850 Volume (Veh/h) simulation HCM 2016 Figure 5: Te efect of volume variations on delay for a single right-turn lane. obtained by simulation was 7.1% higher at the volume of 1,250 Veh/h, while the diference was 23.2% at the volume of Table 3: F-test performed between simulation and HCM2016. 1,450 Veh/h. Method Variance F P value F critical As mentioned in section 4.1, the diference between the Simulation 21.71 simulation and HCM2016 control delay results was verifed 5.99 0.001 2.48 HCM2016 3.62 statistically using a two-sample F-test for variances method in dual right-turn lane. Table 4 shows a signifcant diference between simulation and HCM2016 in control delay results at 4.2. Single Left-Turn Lane. Te through and right-turn a 95% confdence level (P value <0.05). volumes were considered constant for all directions (i.e., 400 and 150 Veh/h). Te left-turn volume varied in the range of 100–400 Veh/h with an incremental step of 50 Veh/h. Fig- 4.4. Dual Left-Turn Lanes. Te through and right-turn ure 6 shows that the intersection delay increased as the left- volumes were considered constant for all directions (i.e., 400 turn volume grew. Tis is true for both simulation and and 150 Veh/h). Te left-turn volume varied between 100 HCM2016 results. Comparing the results of simulation and 700 Veh/h with an incremental step of 50 Veh/h. Fig- software and HCM2016 showed that the intersection delay ure 8 shows that the intersection delay increased as the left- diference was neglectable for the trafc volumes of turn volume grew. Tis is true for both simulation and 100–350 Veh/h. However, the diference became 10% and HCM2016 results. Comparing simulation and HCM2016 34.6% for the volumes of 350 and 400 Veh/h, respectively. results indicated that the intersection delay diference was Tus, the volume of 350 Veh/h was the threshold for a 17% for the trafc volumes of 100–400 Veh/h. However, the sudden change in delay. diference decreased at the range of 450–650 Veh/h, and for the volumes of 550, 600, and 650 Veh/h, the diference became insignifcant. For the trafc volume of 700 Veh/h, 4.3. Dual Right-Turn Lanes. Figure 7 shows the efect of the the diference between the simulation results and HCM2016 number of right-turn lanes on the dual-lane intersection reached 11.4%. delay. Te right-turn volume varied between 200 and 1,450 Veh/h with an incremental step of 50 Veh/h. Te simulation results showed that the delay parameter initially 4.5.ComparisonofSingleandDualLanes. Figure 9 shows the decreased for the trafc volume up to 550 Veh/h. For higher volumes, the delay parameter increased. Tere was a 39.8% efect of the number of right-turn lanes on the intersection diference between the simulation results and HCM2016 at delay estimated by the simulation and HCM2016. As the the volume of 550 Veh/h. HCM2016 results indicated that number of right-turn lanes increased to two, the HCM2016 the delay was independent of the incoming volume below did not estimate any changes in the intersection delay for the 950 Veh/h. Te delay increased as the volume increased from single right-turn lane with a volume below 550 Veh/h and 950 to 1,450 Veh/h. For the trafc volume of 1,200 Veh/h, dual right-turn lanes with a volume below 1,000 Veh/h. On the delay reported by the simulation and HCM2016 was the the other hand, the diference between the delays reported by same. As the volume increased from 1,200 to 1,450 Veh/h, HCM2016 was neglectable for single and dual right-turn lanes at trafc volumes below 500 Veh/h. However, the the growing trend of the delay in the simulation was more than that of HCM2016. For instance, the delay parameter simulation results indicated that the delay was reduced for Delay (s) 8 Journal of Advanced Transportation 50 100 150 200 250 300 350 400 450 Volume (Veh/h) simulation HCM 2016 Figure 6: Te efect of volume variations on delay for single left-turn lane. 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400 1500 Volume (Veh/h) simulation HCM 2016 Figure 7: Te efect of volume variations on delay for dual right-turn lane. 800 Veh/h. At 1,450 Veh/h, the delay was 23 s for the dual Table 4: F-test performed between simulation and HCM2016. right-turn lanes. Te trafc scenarios of signalized intersections were Methods Variance F P value F critical analyzed for diferent V/C values with a maximum of 1. Simulation 19.44 6.04 0.001 1.95 Figure 10 shows the simulation results of the single and dual HCM2016 3.21 right-turn lanes. As the volume increased from 100 to 300 Veh/h, V/C remained fxed at 0.39 for the single right- turn lane. For trafc volumes in the range of 300–800 Veh/h, dual right-turn lanes with volumes in the range of V/C suddenly increased to 0.89. For the dual right-turn 100–550 Veh/h. For higher volumes up to 1,450 Veh/h, the lanes, V/C was constantly equal to 0.39 for the trafc volume delay increased. Tis was the same for the single right-turn of 100 to 600 Veh/h. For higher volumes, V/C gradually lane. Te only diference was that the delay parameter de- increased and reached 0.91 at 1,450 Veh/h. Comparing the creased in the range of 100–300 Veh/h and then increased single and dual right-turn lanes suggested that, similar to the for higher volumes. Te simulation results showed that in delay parameter, V/C converged faster to 1 at low trafc the range of 100–300 Veh/h, the delay associated with the volumes when using a single lane, but for large trafc single and dual lanes was equal. A signifcant diference was volumes, the convergence speed was higher when using dual observed for volumes higher than 300 Veh/h. For a single lanes. Tis is due to the efect of the number of right-turn right-turn lane, the delay was 23 s for the trafc volume of Delay (s) Delay (s) Journal of Advanced Transportation 9 100 200 300 400 500 600 700 Volume (Veh/h) simulation HCM 2016 Figure 8: Te efect of volume variations on delay for dual left-turn lane. 50 250 450 650 850 1050 1250 1450 1650 Volume (Veh/h) Simulation (2-lane) HCM2016 (2-lane) Simulation (1-lane) HCM2016 (1-lane) Figure 9: Delay variations of the efect of changing trafc scenarios in single and dual right-turn lanes. 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 50 150 250 350 450 550 650 750 850 950 1050 1150 1250 1350 1450 Volume (Veh/h) V/C Right-turn 1-lane V/C Right-turn2-lane Figure 10: Comparison of changing a single and dual right-turn lane in volume to capacity (v/c) ratio. Delay (s) Max (V/C) Delay (s) 10 Journal of Advanced Transportation 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 Volume (Veh/h) simulation (1 lane) HCM 2016 (1 lane) simulation (2 lane) HCM 2016 (2 lane) Figure 11: Delay variations of the efect of changing trafc scenarios in a single and dual left-turn lane. 1.2 0.8 0.6 0.4 0.2 0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 Volume (Veh/h) Lef-turn 1-lane Lef-turn 2-lane Figure 12: Comparison of changing single to dual left-turn lane on volume to capacity (v/c) ratio. lanes. For example, for V/C � 0.89, the volume was 800 Veh/ Figure 12 depicts the simulation results of the single and h for a single lane and 1,400 Veh/h for dual lanes (i.e., 75% dual left-turn lanes. As the left-turn volume increased from higher trafc volume). 50 to 100 Veh/h, V/C remained constant at 0.31 for the single Figure 11 illustrates that by increasing the number of left- right-turn lane. For trafc volumes in the range of turn lanes to two, the delay was 20 s for a single left-turn lane at 100–400 Veh/h, V/C increased to 0.98. For the dual right- 350 Veh/h, while the same delay was obtained by the dual left- turn lanes, V/C was constantly equal to 0.28 for the trafc turn lanes at 700 Veh/h, according to HCM2016. Te simulation volume of 50 to 200 Veh/h. For higher volumes, V/C results indicated that increasing the number of left-turn lanes led gradually increased and reached 0.89 at 700 Veh/h. Com- to a delay reduction from 45.6 to 10.3 s for the left-turn volume paring the single and dual left-turn lanes suggested that, of 400 Veh/h. Tis shows that even for smaller volumes (around similar to the delay parameter, V/C converged faster to 1 at 400 Veh/h) in which the signalized intersection was unsaturated, low trafc volumes when using a single lane, but for large there was a signifcant diference in delay parameter between the trafc volumes, the convergence speed was higher when single and dual left-turn lanes in both simulation results and using dual lanes. For example, for V/C � 0.86, the volume HCM, respectively, 77.4% and 59.7%. Terefore, the diference was 350 Veh/h for a single lane and 700 Veh/h for dual lanes can be attributed to the number of lanes. (i.e., 100% higher trafc volume). Max (V/C) Delay (s) Journal of Advanced Transportation 11 depending on the number of right-turn lanes and Table 5: t-test performed between the single and dual lanes. volumes. Furthermore, at low volumes, the delay Turn type Method t stat P value t Critical value decreased and then increased in both single Right- and dual right-turn lanes. Simulation and HCM2016 −3.96 0.001 2.02 turn Left-turn Simulation and HCM2016 −2.21 0.03 2.1 Overall, if the intersection-occupied area is enough, increasing the number of right- and left-turn lanes can be a suitable alternative for at-grade signalized intersections to Te diference between the single and dual lanes (both improve the throughput of an intersection. Our results can right and left turn) results was verifed statistically using the ofer a good vision for trafc engineers to modify the ge- two-sample t-test method. Table 5 shows a signifcant dif- ometry of the four-leg undersaturated signalized intersec- ference between single and dual lanes in volume results at a tions under conditions that the right or left-turn volumes are 95% confdence level (P value <0.05). incremental, and the through volumes are constant. AASHTO 2018 and NCHRP 780 provide recommen- dations based on driving behavior and trafc parameters for Data Availability the length of right- and left-turn lanes in a given condition of geometric elements of at-grade signalized intersections. No datasets were generated or analyzed during the current Also, previous studies conducted in this feld have examined study. the length of the lane in diferent turning movements of an approach (see Table 1). In contrast, the result analysis of this research shows that the number of trafc lanes can signif- Conflicts of Interest icantly impact increasing the throughput of intersections. In Te authors declare that they have no conficts of interest. this study, we have focused on the efect of the number of turning lanes on the volume changes of right turn and left turn. Analyses in undersaturated conditions (V/C< 1) References showed that the HCM2016 delay model is not sensitive to [1] M. E. M. A. Al-Omari and M. Abdel-Aty, “Evaluation of a volume changes, especially in right-turning movements, up New Intersection Design,“Shifting Movements,” Trans- to a specifc value. While much of the research reviewed in portation Research Record, vol. 2675, 2021. the background has been done for V/C � 1 or V/C> 1, this [2] S. Kikuchi, M. Kii, and P. Chakroborty, “Lengths of double or insensitivity of HCM is not seen in saturated and over- dual left-turn lanes,” Transportation Research Record, saturated conditions. vol. 1881, no. 1, pp. 72–78, 2004. [3] M. Liu, L. Xu, L. Shen, and S. Jin, “Modeling capacity at 5. Conclusion signalized intersections with a left-turn storage bay consid- ering signal timing plan,” Journal of Transportation Engi- Tis study sought to assess the performance of signalized neering, Part A: Systems, vol. 145, no. 2, Article ID 04018084, intersections by changing the number of right- and left- 2019. turn approaches and trafc scenarios for a symmetrical [4] S. L. Hallmark and N. Hawkins, “Adding Turn Lanes/ Channelization Tech Brief,” 2014, https://www.intrans.iastate. four-leg intersection using Synchro. For all trafc sce- edu/. narios, the incoming volume of the through approach [5] Y. Guo, P. Liu, Y. Wu, and J. Chen, “Evaluating how right- was considered constant, and the channelized single or turn treatments afect right-turn-on-red conficts at signalized dual right-turn lane approaches and the left-turn volume intersections,” Journal of Transportation Safety and Security, approach with single or dual lane storage length were vol. 12, no. 3, pp. 419–440, 2020. increased under unsaturated trafc conditions. Te fol- [6] E. Nyame-Baaf, C. A. Adams, and K. K. Osei, “Volume lowing conclusions can be drawn: warrants for major and minor roads left-turning trafc lanes at unsignalized T-intersections: a case study using VISSIM (i) As the number of left-turn lanes increased, the re- modelling,” Journal of Trafc and Transportation Engineering, sults showed that for low trafc volumes (less than vol. 5, no. 5, pp. 417–428, 2018. 400 Veh/h) in which the signalized intersection was [7] K. Fitzpatrick, W. H. Schneider, and E. S. Park, “Operation unsaturated, the diference in delay parameter be- and safety of right-turn lane designs,”TransportationResearch tween the single and dual left-turn lanes was sig- Record, vol. 1961, no. 1, pp. 55–64, 2006. nifcant in simulation and HCM2016, respectively, [8] W. Guo, W. Wang, G. Wets et al., “Infuence of stretching- 77.4% and 59.7%. segment storage length on urban trafc fow in signalized intersection,” International Journal of Computational Intel- (ii) Generally, as the through approach volume ligence Systems, vol. 4, no. 6, pp. 1401–1406, 2011. remained constant, the delay parameter reported by [9] S. Farivar, R. Andalibian, and Z. Z. Tian, “Modeling right-turn HCM2016 was fxed up to a certain threshold (10.8 s) blockage and approach capacity at signalized intersections for increased channelized right-turn volume. Te with channelized right-turn lanes,” Journal of Transportation delay parameter of the undersaturated intersection Engineering, Part A: Systems, vol. 143, no. 2, Article ID was less sensitive to single and dual right-turn lane 04016006, 2017. changes. However, the simulation results indicated [10] F. Dion, H. Rakha, and Y. S. Kang, “Comparison of delay that the delay parameter was not constant, estimates at under-saturated and over-saturated pre-timed 12 Journal of Advanced Transportation signalized intersections,” Transportation Research Part B: [27] R. Yao, “Settings of short left-turn lane and signal phase Methodological, vol. 38, no. 2, pp. 99–122, 2004. sequence for isolated signalized intersections,” Transport, vol. 31, no. 4, pp. 416–426, 2014. [11] R. Yao and H. Michael Zhang, “Optimal allocation of lane [28] W. Ma, Y. Liu, J. Zhao, and N. Wu, “Increasing the capacity of space and green splits of isolated signalized intersections with signalized intersections with left-turn waiting areas,” Trans- short left-turn lanes,” Journal of Transportation Engineering, portation Research Part A: Policy and Practice, vol. 105, vol. 139, no. 7, pp. 667–677, 2013. pp. 181–196, 2017. [12] HCM, Highway Capacity Manual. Transportation Research [29] S. Kikuchi and N. Kronprasert, “Determining the length of the Board, National Research Council, Washington, D.C., USA, right-turn lane at a signalized intersection,” Transportation Research Record, vol. 2060, no. 1, pp. 19–28, 2008. [13] AASHTO (American Association of State Highway and [30] G. S. Macfarlane, M. Saito, and G. G. Schultz, “Delay under- Transportation ofcials), A Policy on Geometric Design of estimation at free right-turn channelized intersections,” Pro- Highways and Streets, AASHTO, Washington”, D.C. USA, cedia-social and behavioral sciences, vol. 16, pp. 560–567, 2011. [31] X. Chen, Y. Qi, and D. Li, “Estimating right-turn-on-red capacity [14] F. Wang, K. Li, C. Shao, J. Zhang, B. Li, and N. Han, “Op- for dual right-turn lanes at signalized intersections,” Trans- timisation of signal timing at intersections with waiting portation Research Record, vol. 2286, no. 1, pp. 29–38, 2012. areas,” Promet - Trafc and Transportation, vol. 34, no. 2, [32] X. Chen, Y. Qi, and G. Liu, “Empirical study of gap-accep- pp. 195–209, 2022. tance behavior of right-turn-on-red drivers on dual right-turn [15] R. Guo, J. Liu, and Y. Qi, “An innovative signal timing strategy lanes,” Journal of Transportation Engineering, vol. 139, no. 2, for implementing contrafow left-turn lanes at signalized pp. 173–180, 2013. intersections with split phasing,”Sustainability, vol. 13, no. 11, [33] B. Li, W. Cheng, Y. Bie, and B. Sun, “Capacity of advance p. 6307, 2021. right-turn motorized vehicles at signalized intersections for [16] J. Wu, P. Liu, Z. Z. Tian, and C. Xu, “Operational analysis of mixed trafc conditions,” Mathematical Problems in Engi- the contrafow left-turn lane design at signalized intersections neering, vol. 2019, Article ID 3854604, 13 pages, 2019. in China,” Transportation Research Part C: Emerging Tech- [34] S. Kikuchi, N. Kronprasert, and M. Kii, “Lengths of turn lanes nologies, vol. 69, pp. 228–241, 2016. on intersection approaches: three-branch fork lanes–left-turn, [17] Z. Li and L. Elefteriadou, “Maximizing the trafc throughput through, and right-turn lanes,” Transportation Research Re- cord, vol. 2023, no. 1, pp. 92–101, 2007. of turn bays at a signalized intersection approach,” Journal of [35] D. Zupanovi´ c, M. Anzek, ˇ and G. Kos, “Optimisation of signal- Transportation Engineering, vol. 139, no. 5, pp. 425–432, 2013. controlled intersection capacity,” Promet - Trafc and [18] Y. Zheng, X. Hua, W. Wang, J. Xiao, and D. Li, “Analysis of a Transportation, vol. 22, no. 6, pp. 419–431, 2012. signalized intersection with dynamic use of the left-turn lane [36] W. Wu, W. Ma, and K. Long, “Capacity matching based for opposite through trafc,” Sustainability, vol. 12, no. 18, model for protected left turn phases design of adjacent sig- p. 7530, 2020. nalized intersections along arterial roads,” Promet - Trafc [19] I. Shatnawi, P. Yi, and I. Khliefat, “Automated intersection and Transportation, vol. 27, no. 1, pp. 13–21, 2015. delay estimation using the input–output principle and turning [37] H. K. Kim, S. K. Oh, and G. M. Bae, “Demonstrative com- movement data,” International Journal of Transportation parison of two trafc signal optimization programs: synchro-6 Science and Technology, vol. 7, no. 2, pp. 137–150, 2018. and PASSER II,” KSCE Journal of Civil Engineering, vol. 10, [20] A. Abdi Kordani, O. Rahmani, A. S. Abdollahzadeh Nasiri, no. 1, pp. 41–45, 2006. and S. M. Boroomandrad, “Efect of adverse weather con- [38] H. Xu, H. Liu, and Z. Tian, “Control delay at signalized diamond ditions on vehicle braking distance of highways,” Civil En- interchanges considering internal queue spillback,” Trans- gineering Journal, vol. 4, no. 1, pp. 46–57, 2018. portation Research Record, vol. 2173, no. 1, pp. 123–132, 2010. [21] Y. Qi, L. Yu, M. Azimi, and L. Guo, “Determination of storage [39] C. Zheng, P. Liu, J. J. Lu, and H. Chen, “Evaluating the efect lengths of left-turn lanes at signalized intersections,” Trans- efects of U-turns on level of service of signalized intersections portation Research Record, vol. 2023, no. 1, pp. 102–111, 2007. using synchro and SimTrafc,” in Proceedingsofthe2009IEEE [22] K. Yin, Y. Zhang, and B. X. Wang, “Analytical models for Intelligent Vehicles Symposium, pp. 971–976, IEEE, Xian, protected plus permitted left-turn capacity at signalized in- China, June 2009. tersection with heavy trafc,” Transportation Research Record, [40] J. Zheng, H. X. Liu, S. Misgen, K. Schwartz, B. Green, and vol. 2192, no. 1, pp. 177–184, 2010. M. Anderson, “Use of event-based trafc data in generating [23] S. Kikuchi and N. Kronprasert, “Determining lengths of left- time–space diagrams for evaluation of signal coordination,” turn lanes at signalized intersections under diferent left-turn TransportationResearchRecord, vol. 2439, no. 1, pp. 94–104, 2014. signal schemes,” Transportation Research Record, vol. 2195, [41] H. Hatami and I. Aghayan, “Trafc efciency evaluation of elliptical roundabout compared with modern and turbo no. 1, pp. 70–81, 2010. roundabouts considering trafc signal control,” Promet - [24] W. L. Reynolds, X. Zhou, N. M. Rouphail, and M. Li, “Es- Trafc and Transportation, vol. 29, no. 1, pp. 1–11, 2017. timating sustained service rates at signalized intersections [42] A. R. Dabiri, I. Aghayan, and F. Hadadi, “A comparative with short left-turn pockets: mesoscopic approach,” Trans- analysis of the performance of turbo roundabouts based on portation Research Record, vol. 2173, no. 1, pp. 64–71, 2010. geometric characteristics and trafc scenarios,” Trans- [25] K. Yin, Y. Zhang, and B. X. Wang, “Modeling delay during portation letters, vol. 13, no. 9, pp. 674–685, 2021. heavy trafc for signalized intersections with short left-turn [43] K. Fitzpatrick, M. A. Brewer, P. Dorothy, and E. S. Park, bays,” Transportation Research Record, vol. 2257, no. 1, “Design Guidance For Intersection Auxiliary Lanes (No. pp. 103–110, 2011. Project 03-102,” NCHRP 780, 2014, https://www.trb.org/ [26] Y. G. Qi, L. Guo, L. Yu, and H. Teng, “Estimation of design Publications/Blurbs/171328.aspx., 2014. lengths of left-turn lanes,” Journal of Transportation Engi- neering, vol. 138, no. 3, pp. 274–283, 2012.

Journal

Journal of Advanced Transportation – Hindawi Publishing Corporation

Published: Feb 16, 2023

There are no references for this article.