1 Introduction
2 Literature review
Index of study | Composition of mixed traffic | Method | Performance measures | Discussed variables | Main results |
---|---|---|---|---|---|
Link-based studies | |||||
1 Hussain et al. [7] | CAV and HV | An analytical model | Capacity; Throughput | MPRs; Mixed headway settings; Number of CAV DLs; Mixed traffic demand | More aggressive CAVs need less DLs |
2 Ghiasi et al. [1] | CAV and HV | A Markov chain method | Capacity | MPRs; CAV platooning intensity; Mixed headway settings; Number of CAV DLs | No CAV DL is the optimal solution in the unsaturated traffic or when CAV adopts a larger headway than that of mixed traffic; The number of CAV DL should be increased with the CAV demand |
3 Chen et al. [6] | AV (automated vehicle) and HV | A theoretical framework of capacity | Capacity | MPRs; Demands; Segregation policy | The strict segregation of AVs and HVs will lead to the lower capacity; AVs should be distributed to the most efficient lanes |
4 Ramezani et al. [8] | AV and HV | A theoretical model for headway, capacity and delay | Capacity; Delay | MPRs; Mixed headway settings; Multiple lane configurations | The achievement of minimum delay depends on the MPR |
5 Ivanchev et al. [9] | AV and HV | An analytical evaluation based on simulation | Throughput; BPR-based travel time; Fuel consumption | MPRs; With or without a DL; Headways of AV | Travel times of AVs are significantly reduced with a low MPR; HVs are delayed due to the reduced capacity for them |
6 Talebpour et al. [10] | AV and HV | Simulation | Throughput; Travel time reliability | MPRs; Three DL access strategies | The optional use of DL can relieve the congestion; The potential benefit of DL to throughput can be seen when MPR is over 50% for the 2-lane highway and 30% for the 4-lane highway |
7 Laan and Sadabadi [11] | AV and HV | CORridor MACro simulation | Flow; Speed | MPRs; Reaction times of AV | The performance of implementing a DL increases with MPRs adding up to 30%, 40%, or 50%, and then, it will deteriorate |
8 Ye and Yamamoto [12] | CAV and HV | Simulation based on cellular automation | Flow; Flow-density diagram | MPRs; Demand levels; Number of DLs; Speed limit | When MPR is low, a CAV DL will deteriorate the traffic throughput; The benefit of DLs could be obtained within a medium density range; A higher speed limit for CAVs on a DL is beneficial |
9 Abdel-Aty et al. [13] | CV (connected vehicle) and HV | Vissim simulation | SSAM; Average speed; Average delay | MPRs; Multiple lane configurations | Managed CV platooning lane could significantly improve the traffic speed |
10 Nickkar and Lee [14] | AV and HV | AIMSUM and SIDRA simulation | Travel time; Delay; Speeds; Queue | MPRs; With or without AV DLs | The improvement brought by the DLs to the performance of a roundabout can be seen at a high MPR, but it is not significant |
11 Wang et al. [15] | CAV and HV | Simulation | Capacity | MPRs CAV platoon coefficient | The impact on the on-ramp junctions |
Network-based studies | |||||
13 Chen et al. [16] | AV and HV | A multi-class network equilibrium model | Minimize the social costs | MPRs (Market Penetration Rates); The deployment plan of AV DLs | A progressive deployment of AV lanes; Wide deployment when MPR reaches a high level (e.g., 20%) |
14 Qom et al. [17] | CACC (Cooperative Adaptive Cruise Control) vehicle and HV | Static and dynamic traffic assignment model | Throughput | MPRs; A tolling policy; Traffic demand | Results from STA (Static Traffic Assignment) and DTA (Dynamic Traffic Assignment) are consistent; The toll incentives are not beneficial until the MPR reaches a high level |
15 Liu and Song [18] | CAV and HV | User equilibrium model | Capacity; Travel time | MPRs; With or without CAV DLs; CAV DL toll rates | The implementation of AV lanes or AV toll lanes can significantly improve the traffic performance |
16 Wang et al. [15] | CAV and HV | A multiclass traffic assignment model with elastic demand | Link flow | CAV DL toll rates; Traffic demands | The optimal toll rates for the HVs using CAV DL |
Index of study | Automation level of CAV | Data source | Headway between HVs (s) | Headway for an HV following a CAV (s) | Headway for a CAV following an HV (s) | Headway between CAVs (s) |
---|---|---|---|---|---|---|
1 Joel VanderWerf et al. [20] | ACC | Simulation | – | – | Uniform 1.0–2.0 | Uniform 0.5–1.4 |
2 Bose and Ioannou, [21] | ACC/CACC | Simulation | Uniform 0.7–2.2 | – | Uniform 0.5–1.5 | – |
3 Arem et al. [22] | CACC | Simulation | Fixed 1.4 | Fixed 1.4 | Fixed 1.4 | Fixed 0.5 |
4 Christopher Nowakowski et al. [23] | CACC | Field test | – | – | Uniform 1.1–2.2 | Uniform 0.6–1.1 |
5 Schakel et al. [24] | ACC | Field test | – | – | Gaussian 1.2 ± 0.15 and 1.2 ± 0.3 | Gaussian 1.2 ± 0.15 and 1.2 ± 0.3 |
6 Larsson [25] | ACC | Survey | – | – | Uniform 1.0–2.6 | |
7 Altay et al. [26] | ACC/CACC | Field test | – | – | Uniform 0.6–2.0 | Uniform 0.6–2.0 |
8 Li Zhao and Sun [27] | ACC/CACC | Simulation | – | – | Fixed 1.4 | Fixed 0.5 |
9 Kumar et al. [28] | None | Field data | Uniform 1.3–2.4 | – | ||
10 Arnaout and Bowling [29] | CACC | Simulation | Uniform 1.0–1.8 | Uniform 1.0–1.8 | Uniform 0.8–1.0 | Fixed 0.5 |
11 Nikolos et al. [30] | ACC/CACC | Numerical Simulations | – | – | Uniform 0.8–2.2 | – |
12 Hussain et al. [7] | CAV | Numerical analysis | Fixed 1.8 | 1.2 or 1.8 | 1.2 or 1.8 | 0.3 and 0.45 |
13 Ivanchev et al. [9] | AV | Agent-based macroscopic simulation | Fixed 1.8 | Fixed 1.8 | Uniform 0.5–1.0 | Uniform 0.5–1.0 |
14 Darbha et al. [31] | ACC | Numerical simulation | – | – | – | 0.68–0.88 or 0.27–0.5 or 0.31–0.58 |
15 [1] | CAV | Numerical analysis | Uniform 0.8–2.2 | Uniform 0.8–2.2 | Uniform 0.7–1.5 | Uniform 0.6–1.1 |
16 [32] | AV | Simulation | Fixed 1.8 | Fixed 1.8 | Fixed 1.2 | Fixed 0.9 |
17 Lu et al. [33] | AV | Simulation | Fixed 0.9 | Fixed 0.9 | Fixed 0.6 | Fixed 0.6 |
18 Nishimura et al. [34] | AV | Simulation | Normal average 1.69 | Normal average 1.69 | Uniform 0.1–3.0 | Uniform 0.1–3.0 |
19 Zhao et al. [35] | CAV | Simulation | Fixed 1.2 | Fixed 1.2 | Fixed 0.9 | Fixed 0.9 |
20 Martin-Gasulla et al. [19] | CAV | Simulation | Fixed 0.9 | Fixed 0.9 | Uniform 1.5–2.5 | Fixed 0.6 |
3 Methodology
3.1 Dedicated lane policies
3.2 Efficiency measures
3.2.1 Capacity
3.2.2 Throughput
4 Numerical analysis
4.1 Experiment settings
Number of lanes | Number of DL | DL accessibility | Index of scenario | Lane configurations |
---|---|---|---|---|
2 | 0 | – | 2–1 | 2 MLs |
1 | Mandatory | 2–2 | 1 DL and 1 GPL | |
Optional | 2–3 | 1 DL and 1 ML | ||
3 | 0 | – | 3–1 | 3 MLs |
1 | Mandatory | 3–2 | 1 DL and 2 GPLs | |
Optional | 3–3 | 1 DL and 2 MLs | ||
2 | Mandatory | 3–4 | 2 DLs and 1 GPL | |
Optional | 3–5 | 2 DLs and 1 ML | ||
4 | 0 | / | 4–1 | 4 MLs |
1 | Mandatory | 4–2 | 1 DL and 3 GPLs | |
Optional | 4–3 | 1 DL and 3 MLs | ||
2 | Mandatory | 4–4 | 2 DLs and 2 GPLs | |
Optional | 4–5 | 2 DLs and 2 MLs | ||
3 | Mandatory | 4–6 | 3 DLs and 1 GPL | |
Optional | 4–7 | 3 DLs and 1 ML |
4.2 Results and evaluations
4.2.1 Capacity
Index of scenario | Capacity under different headway distributions (veh/h) | Ideal condition (the value of \({P}_{CAV}\)) for aggressive, neutral, conservative, and safe driving modes respectively | |||
---|---|---|---|---|---|
Aggressive | Neutral | Conservative | Safe | ||
2–1 | [3614, 9000] | [3610, 7200] | [3604, 4800] | [3272, 4800] | 100% |
2–2 | 6300 | 5400 | 4200 | 4200 | 71%, 67%, 57% and 57% |
2–3 | [6307, 9000] | [5405, 7200] | [4202, 4800] | [4036, 4800] | 100% |
3–1 | [5321, 13,500] | [5415, 10,800] | [5406, 7200] | [4908, 7200] | 100% |
3–2 | 8100 | 7200 | 6000 | 6000 | 56%, 50%, 40% and 40% |
3–3 | [8114, 13,500] | [7210, 10800] | [6004, 7200] | [5672, 7200] | 100% |
3–4 | 10,800 | 9000 | 6600 | 6600 | 83%, 80%, 73% and 73% |
3–5 | [10807, 13,500] | [9005, 10,800] | [6602, 7200] | [6436, 7200] | 100% |
4–1 | [7228, 18,000] | [7220, 14,400] | [7208, 9600] | [6544, 9600] | 100% |
4–2 | 9900 | 9000 | 7800 | 7800 | 45%, 40%, 31% and 31% |
4–3 | [9921, 18,000] | [9015, 14,400] | [7806, 9600] | [7308, 9600] | 100% |
4–4 | 12,600 | 10,800 | 8400 | 8400 | 71%, 67%, 57% and 57% |
4–5 | [12614, 18,000] | [10810, 14,400] | [8404, 9600] | [8072, 9600] | 100% |
4–6 | 15,300 | 12,600 | 9000 | 9000 | 88%, 85%, 80% and 80% |
4–7 | [15307, 18,000] | [12605, 14,400] | [9002, 9600] | [8836, 9600] | 100% |