design of an efficient vehicle-actuated signal control...
TRANSCRIPT
![Page 1: Design of an Efficient Vehicle-Actuated Signal Control ...scientiairanica.sharif.edu/article_22418_3cb77f0f300eb1046401b00b… · 1 1 Design of an Efficient Vehicle-Actuated Signal](https://reader035.vdocuments.mx/reader035/viewer/2022070221/61357ca30ad5d206764768b1/html5/thumbnails/1.jpg)
1
Design of an Efficient Vehicle-Actuated Signal Control Logic for Signalized 1
Intersections 2
3
Ziya CAKICIa,*, Yetis Sazi MURATb, Metin Mutlu AYDINc 4
a Bayburt University, Faculty of Engineering, Department of Civil Engineering, Bayburt 69000, Turkey 5
b Pamukkale University, Faculty of Engineering, Department of Civil Engineering, Denizli 20160, Turkey 6
c Ondokuz Mayıs University, Faculty of Engineering, Department of Civil Engineering, Samsun 55000, Turkey 7
8
* Corresponding author: 9
E-mail: [email protected]; Phone no: +90 554 204 5924 10
11
Abstract 12
The effectiveness of fixed-time management systems dramatically decreases in case of 13
fluctuated traffic demands at signalized intersection approaches. This leads to the waste of time 14
in traffic and may cause material, psychological and ecological problems. Especially in recent 15
years, to minimize the negative impacts of these problems, many researchers focus on 16
Intelligent Transportation Systems (ITS). As it is known, one of the Intelligent Transportation 17
Systems applications is called vehicle-actuated (traffic-actuated) management systems. 18
Because the success of these types of applications is directly related to the created control logic, 19
the selection of the most proper control parameters for the created control logic is an important 20
issue. This study aims to create an effective control logic and flow chart for the vehicle-actuated 21
management system. At the end of the analyses, it is seen that the created vehicle-actuated 22
management system can adapt to fluctuations in traffic demands at signalized intersection 23
approaches. Average vehicle delays, fuel consumptions and exhaust emissions can be reduced 24
significantly by the created system. Especially, in case of fluctuations in traffic demands at 25
![Page 2: Design of an Efficient Vehicle-Actuated Signal Control ...scientiairanica.sharif.edu/article_22418_3cb77f0f300eb1046401b00b… · 1 1 Design of an Efficient Vehicle-Actuated Signal](https://reader035.vdocuments.mx/reader035/viewer/2022070221/61357ca30ad5d206764768b1/html5/thumbnails/2.jpg)
2
intersection approaches exist, it is concluded that the performance of the intersection can 26
increase enormously with the created vehicle-actuated management system. 27
Keywords: Fluctuated traffic demand; Vehicle-actuated management, Delay, Fixed-time 28
management, Simulation. 29
1. Introduction 30
Economic and technological developments all over the world make life easier. However, this 31
situation may cause several negative effects in many areas. Highways are undoubtedly one of 32
the areas where these problems are mostly seen. In parallel with the increase in demand for 33
passenger and freight transportation, the number of vehicles has increased. This increase can 34
cause many economic and environmental problems [1-3]. To minimize or eliminate the 35
negative effects of these problems on humans and the environment, geometric-topographic and 36
operational arrangements at road networks can be pointed out as important parts of 37
environment-economy-friendly measures and reasonable solutions. At this stage, scientific-38
based analysis and solution approaches should be preferred. Minimizing the stop numbers and 39
delays is an important step to solve this problem at signalized intersections [4]. The 40
minimization of delays and stops at signalized intersections can be achieved in two different 41
ways: with most proper designing of intersection geometry (geometric and topographic 42
arrangements) and with determining of signal timings considering traffic demands at 43
intersection approaches (operational arrangement) [5-8]. Hence, effective signal management 44
is a quite important factor to reduce the delay, fuel consumption, noise pollution and exhaust 45
emissions and to increase the intersection capacity and level of service of intersection [9-12]. 46
In the absence of the fluctuations in traffic demands at intersection approaches, the optimum 47
fixed-time management system, which is widely used today, maybe the right and appropriate 48
choice to manage the traffic flows at intersections [13]. But sometimes sports events, concerts, 49
traffic accidents, bad weather conditions can cause short or long-term fluctuations in traffic 50
![Page 3: Design of an Efficient Vehicle-Actuated Signal Control ...scientiairanica.sharif.edu/article_22418_3cb77f0f300eb1046401b00b… · 1 1 Design of an Efficient Vehicle-Actuated Signal](https://reader035.vdocuments.mx/reader035/viewer/2022070221/61357ca30ad5d206764768b1/html5/thumbnails/3.jpg)
3
demands at some intersection approaches. In such cases, the operating efficiency of the 51
optimum fixed-time management system decreases especially at intersection approaches where 52
the fluctuations in traffic demands exist. This situation may cause long-term delays, long 53
queues, excessive fuel consumption and exhaust emissions [13-16]. It has been demonstrated 54
in many studies by researchers that in the event of fluctuations in traffic demands, optimum 55
fixed-time signal management systems cannot provide efficient results [17, 18]. 56
For the last 60 years, a lot of studies have been conducted on signal control systems. And, the 57
studies have continued today to improve their performances [19-21]. Since the last quarter of 58
the 20th century, it has been also aimed to make these systems more effective for high traffic 59
demands at signalized intersections and thus it has been aimed to increase the operational 60
efficiency further [22-26]. Especially in recent years, there has been a significant increase in 61
the number of studies on the signal management systems and control algorithms that take the 62
instantaneous traffic flows into account. These systems may appear in different ways either 63
optimization or created control algorithm based [27, 28]. 64
Traffic management systems that are operated based on any created control algorithm are 65
referred to as vehicle-actuated (traffic-actuated) signal management systems. Vehicle-actuated 66
management systems determine the green and red signal timings according to the current 67
dynamics of vehicle arrivals and queuing information obtained by detectors placed on the lanes 68
at intersection approaches [28]. Because vehicle-actuated signal management systems generally 69
have positive effects on the intersection performance, many researchers have worked on this 70
issue [26]. Some of these studies can be summarized as follows: 71
Akçelik [22] proposed an approach to estimate the average green times and cycle time for 72
vehicle-actuated management systems. He made an evaluation on the prediction of vehicle 73
arrival headways and tried to determine the optimum cycle time. Trabia et al. [29] worked on 74
fuzzy logic-based vehicle-actuated management systems. As a result of the study, they 75
![Page 4: Design of an Efficient Vehicle-Actuated Signal Control ...scientiairanica.sharif.edu/article_22418_3cb77f0f300eb1046401b00b… · 1 1 Design of an Efficient Vehicle-Actuated Signal](https://reader035.vdocuments.mx/reader035/viewer/2022070221/61357ca30ad5d206764768b1/html5/thumbnails/4.jpg)
4
determined that delays can be reduced by 10% with a fuzzy-based approach. Kim and Courage 76
[30] proposed a maximum green time design method for a vehicle-actuated management system 77
that minimizes delay. Guo et al. [31] developed the two analytical models to optimize the 78
maximum green times which are applied on vehicle-actuated management systems. They 79
specified that these models could adjust maximum green time systematically and cyclically by 80
adapting to the changes in traffic demands at intersections. Zheng et al. [32] worked on a control 81
model which aims to provide adaptive functionality of vehicle-actuated management systems. 82
As a result of the study, it was found that the model has a great potential to improve the 83
performance of the signalized network. Jiang et al. [33] developed a method to optimize the 84
parameters of the vehicle-actuated signal management systems. In another study, Grether et al. 85
[13] evaluated the vehicle-actuated signal control strategy. As a result, they concluded that 86
travel times can be significantly reduced with the vehicle-actuated management system, 87
especially in the event of excessive increases in traffic demands. Swaminathan et al. [34] 88
developed a vehicle-actuated management model. According to analysis results, it was 89
determined that the average delays can be reduced by 28% compared to the current situation 90
with the developed control model. In a similar study, Guo and Ma [35] aimed to develop a new 91
vehicle-actuated signal control model, which resulted in lower average vehicle delays compared 92
to both fixed-time and standard actuated control systems. Ribeiro and Simoes [36] used global 93
optimization and complementarity to determine the green and red signal timings at each cycle 94
for vehicle-actuated management systems. At the end of the analysis, it was concluded that the 95
suggested methodology can be used for the determination of the green and red signal timings, 96
as well as cycle lengths. Lee and Maleki [37] studied on maximum green time settings for 97
vehicle-actuated signal control at isolated intersections. They aimed to determine maximum 98
green times using fuzzy logic approach, dynamically. As a result, it was determined that fuzzy 99
logic is a useful approach to overcome the difficulty of setting suitable maximum green time 100
![Page 5: Design of an Efficient Vehicle-Actuated Signal Control ...scientiairanica.sharif.edu/article_22418_3cb77f0f300eb1046401b00b… · 1 1 Design of an Efficient Vehicle-Actuated Signal](https://reader035.vdocuments.mx/reader035/viewer/2022070221/61357ca30ad5d206764768b1/html5/thumbnails/5.jpg)
5
for any phase. In a different study, Wang et al. [38] investigated the green time extension for 101
vehicle-actuated signal control. As a result, they found that the optimum critical headway value, 102
which is an important parameter for green time extension, decreases with an increase in traffic 103
demand and in some cases, this value may be more than 2-3 seconds as a common assumption. 104
Promraksa et al. [39] studied on vehicle-actuated signal management system for reducing CO2 105
emission. At the end of the analyses, it was concluded that a fully-actuated management system 106
can provide better and more encouraging results than the fixed-time and semi-actuated 107
management systems. 108
As can be seen from the previous studies, the selection of the most proper design parameters 109
(such as minimum critical headway, placement of detectors, minimum green signal timings, 110
maximum green signal timings) for the vehicle-actuated management systems has a direct and 111
significant effect on the system performance. This study guides for designing an efficient 112
vehicle-actuated signal control logic for signalized intersections. In the scope of the study, the 113
selection of the most proper design parameters was evaluated, separately. Then, a new vehicle-114
actuated management system for a four-leg signalized intersection was developed in VISSIM 115
Vehicle Actuated Programming (VISVAP). The efficiency of the developed system was tested 116
on 20 different traffic scenarios considering the different performance criteria (vehicle delay, 117
fuel consumption and exhaust emission) in VISSIM. To determine the success of the developed 118
system, unlike the current studies in the literature, short and long-term fluctuations in traffic 119
demands at the signalized intersection approaches were also taken into account in the analysis. 120
It is seen that the developed system can adapt to the fluctuations in traffic demands and 121
enormously reduce the average vehicle delay, fuel consumption and exhaust emission at the 122
signalized intersection. Thus, it is determined that the developed system can be used for 123
improving the performance of signalized intersections. 124
2. Method 125
![Page 6: Design of an Efficient Vehicle-Actuated Signal Control ...scientiairanica.sharif.edu/article_22418_3cb77f0f300eb1046401b00b… · 1 1 Design of an Efficient Vehicle-Actuated Signal](https://reader035.vdocuments.mx/reader035/viewer/2022070221/61357ca30ad5d206764768b1/html5/thumbnails/6.jpg)
6
2.1.Vehicle-Actuated Management Model for Four-Leg Signalized Intersections: VAM4 126
In vehicle-actuated management, traffic flows at an intersection are managed using data and 127
information from the detectors which are located at the intersection approaches [18]. Vehicle-128
actuated management systems can be classified in two ways as semi-actuated and fully-129
actuated. In semi-actuated management, only one or a few of the intersection approaches have 130
detectors. In fully-actuated management, all intersection approaches have detectors and traffic 131
flow data and information are provided from all approaches. In this type of management, the 132
green signal timings and the cycle times are determined by considering the traffic demands 133
detected. In fully-actuated management systems, the order of phases can be fixed or flexible. 134
Furthermore, if there is no demand for a phase, that phase can be skipped directly. In vehicle-135
actuated management systems, the effective operation of an intersection generally depends on 136
the following factors: minimum green times (sec.), critical arrival headways of traffic flow, 137
placement of detectors (detector locations), maximum green times (sec.). Therefore, choosing 138
reasonable and appropriate values for these factors is crucial to improve signalized 139
intersections’ performance [14, 35]. In this study, the performance of the vehicle-actuated 140
signal management system (VAM4) developed for a four-leg signalized intersection which 141
includes scenarios, where short or long-term fluctuations exist in traffic demands at intersection 142
approaches, was examined. For this purpose, firstly, the control parameters and criteria related 143
to the control algorithm of the proposed VAM4 were determined. In the analysis, considering 144
the high traffic volumes, the minimum critical arrival headway is selected as approximately 2 145
seconds. In this case, as long as the critical arrival headway determined by the detectors does 146
not exceed 2 seconds, the green signal timing of that phase is extended by 2 seconds. When the 147
current studies in the literature are examined, it can be seen that this value is generally selected 148
between 1.5 and 4 seconds. [14, 40]. Therefore, the selected green extension value (2 seconds) 149
was appropriate for the analysis. Besides, when the previous studies are examined carefully, 150
![Page 7: Design of an Efficient Vehicle-Actuated Signal Control ...scientiairanica.sharif.edu/article_22418_3cb77f0f300eb1046401b00b… · 1 1 Design of an Efficient Vehicle-Actuated Signal](https://reader035.vdocuments.mx/reader035/viewer/2022070221/61357ca30ad5d206764768b1/html5/thumbnails/7.jpg)
7
there is no clear and precise information about the placement of detectors for vehicle-actuated 151
management systems. However, the placement of the detectors was directly related to the traffic 152
volume as a common practice in the literature [14, 15, 40]. According to previous studies, the 153
detectors were placed generally 15 m to 45 m before the stop lines. However, there is no any 154
precise information about the placement of the detectors in the literature. In this study, the 155
locations of the detectors were determined by trial and error method for each traffic scenario. 156
According to the results obtained by trial and error, the relationship between the distance of the 157
detectors to the stop line at the intersection approaches and the average delay was determined 158
as given in Fig. 1. This relationship is similar to the relationship depicted by Bullen [14]. 159
When the previous studies are examined, the minimum green time is generally selected as 4-5 160
seconds or above depending on the distance of the detector to the stop line [14, 15]. As 161
mentioned before, especially when the traffic volumes at intersection approaches are high, the 162
distance of the detectors to the stop line can be up to 45 meters. For this reason, the minimum 163
green time value must be determined considering locations (placements) of the detectors at the 164
intersection approaches. Determined minimum green time must allow to clearing of the queues 165
at intersection approaches. Depending on the locations of the detectors, the minimum green 166
times for the vehicle-actuated management systems can be calculated by using Eq. (1) [15]: 167
min L
lG t h integer
e
= +
(1) 168
Where; 169
minG : Minimum green time (sec), 170
Lt : Assumed start-up lost time (sec), 171
h : Assumed saturation headway (sec), 172
l : Distance between detectors and stop line (m), 173
e : Distance between storaged vehicles (average 6-7 m). 174
![Page 8: Design of an Efficient Vehicle-Actuated Signal Control ...scientiairanica.sharif.edu/article_22418_3cb77f0f300eb1046401b00b… · 1 1 Design of an Efficient Vehicle-Actuated Signal](https://reader035.vdocuments.mx/reader035/viewer/2022070221/61357ca30ad5d206764768b1/html5/thumbnails/8.jpg)
8
In the scope of this study, minimum green times are computed by using Eq. (1). Depending on 175
the distance between detectors and stop line, the obtained minimum green times are presented 176
in Table 1. 177
As can be seen from the previous studies, the maximum green time for vehicle-actuated 178
management systems is generally selected between 45 and 60 seconds [41]. In this study, the 179
maximum green time of each phase is selected as 60 seconds and this value is used in the 180
analysis. General information about the proposed VAM4 for the four-leg signalized intersection 181
model is shown in Fig. 2. 182
As can be seen from Fig. 2, the modeled four-leg signalized intersection has a total of 14 183
detectors for 14 lanes (one detector for each lane). These detectors are used for both detecting 184
the queuing at intersection approaches and determining the vehicle arrival headways. In the 185
developed VAM4 model, it is assumed that the modeled four-leg signalized intersection is 186
operated with four phases. Right of way for each intersection approach is provided in separate 187
phases. Right of ways for the West, North, East and South intersection approaches are provided 188
in Phase I, Phase II, Phase III and Phase IV, respectively. Besides, in the developed VAM4 189
model, when excitation is not detected from the detectors which are at one of the intersection 190
approaches, phase transition can be actualized (the order of phases can be changed). The 191
simulation period for the analysis is determined as 3600 sec. The flow chart of the developed 192
VAM4 control logic is presented in Fig. 3. 193
2.2.Traffic Scenarios and Fluctuations in Traffic Demands 194
In the scope of this study, firstly, a movement direction-based reference demand matrix has 195
been created considering the four-leg signalized intersection model shown in Fig. 2. This 196
created reference demand matrix for the analysis is given in Table 2. 197
In the next step, a total of 20 different traffic scenarios are created by using the reference 198
demand matrix given in Table 2 for the modeled four-leg signalized intersection. For each 199
![Page 9: Design of an Efficient Vehicle-Actuated Signal Control ...scientiairanica.sharif.edu/article_22418_3cb77f0f300eb1046401b00b… · 1 1 Design of an Efficient Vehicle-Actuated Signal](https://reader035.vdocuments.mx/reader035/viewer/2022070221/61357ca30ad5d206764768b1/html5/thumbnails/9.jpg)
9
created scenario, movement direction-based traffic volumes showed an increase or decrease 200
based on the reference demand matrix. Besides, some of the movement direction-based traffic 201
volumes are constant in some scenarios. The increase and decrease rates of the movement 202
direction-based traffic volumes for all scenarios are shown in Fig. 4. 203
As can be seen from Fig. 4, when the scenario number increases, general increase trends are 204
seen in movement direction-based traffic volumes. Therefore, the total traffic volumes at the 205
intersection increase when the scenario number increases as shown in Fig. 5. As can be seen in 206
Fig. 5, created traffic scenarios are divided into four groups considering total traffic volumes at 207
the intersection. According to their demand levels, these groups are named as low, moderate, 208
high and very high. When the hourly total traffic demand is between 2000 veh. and 2500 veh., 209
the demand is considered as low. If the hourly total traffic demand is between 2500 veh. and 210
3000 veh., it can be said that the demand is moderate. When the hourly total traffic demand is 211
between 3000 veh. and 3500 veh., the demand is regarded as high. Finally, if the hourly total 212
traffic demand is between 3500 veh. and 4000 veh., it can be said that the demand is very high. 213
Thus, the performance of the VAM4 model developed for a four-leg signalized intersection 214
could be evaluated for the cases of low, moderate, high and very high traffic demands, 215
separately. 216
As it is known, in the VISSIM simulation program, the time intervals when the fluctuations in 217
traffic demands occur and the quantities (proportions) of the fluctuations in these time intervals 218
can be determined by the users. Thus, as a result of the short or long-term fluctuations that may 219
be seen due to the reasons such as concerts, sports events, traffic accidents, adverse weather 220
conditions, changes that occur in the performance of intersection can be analyzed in VISSIM. 221
As mentioned above, in this study, it is aimed to examine the performance and effectiveness of 222
the developed VAM4 in case of fluctuations with variable quantities in variable time intervals 223
for traffic demands at different intersection approaches. For this aim, sample cases related to 224
![Page 10: Design of an Efficient Vehicle-Actuated Signal Control ...scientiairanica.sharif.edu/article_22418_3cb77f0f300eb1046401b00b… · 1 1 Design of an Efficient Vehicle-Actuated Signal](https://reader035.vdocuments.mx/reader035/viewer/2022070221/61357ca30ad5d206764768b1/html5/thumbnails/10.jpg)
10
the quantities of the fluctuations in traffic demands and the time intervals when the fluctuations 225
occur in traffic demands are created for each traffic scenario. The created sample cases are 226
presented in Fig. 6. In Fig. 6, for each traffic scenario, the time intervals when the fluctuations 227
occur in traffic demands at different intersection approaches and the quantities of fluctuations 228
during 60 minutes of the analysis period can be seen in detail. As an example; for Scenario 6, 229
it is assumed that the fluctuations in traffic demands occur twice at the West and North 230
approaches and only once at the East approach during 60 minutes of the analysis period. 231
Besides, for Scenario 6, the following matters can be said about the quantities of fluctuations, 232
and the time intervals when the fluctuations occur in traffic demands: 233
• It is assumed that approximately 33% of the vehicles at the west approach arrive to the 234
intersection between the 20th and 33rd minutes, and approximately 28% arrive to the 235
intersection between the 47th and 60th minutes. 236
• It is assumed that approximately 22% of the vehicles at the north approach arrive to the 237
intersection between the 20th and 27th minutes and 22% arrive to the intersection between 238
the 40th and 47th minutes. 239
• It is assumed that approximately 50% of the vehicles at the east approach arrive to the 240
intersection between the 13th and 33rd minutes. 241
3. Analysis 242
Webster, Highway Capacity Manual (HCM) and Akcelik methods are the most used delay 243
calculation approaches for predicting the average vehicle delay which occurs at signalized 244
intersections [20, 40, 42]. While the Webster and HCM methods are based on the phase-related 245
design, the Akcelik method is based on the movement-related design. In the scope of the study, 246
when the modeled intersection was examined carefully, it was seen that all lanes at intersection 247
approaches were designed based on the traffic flows’ movement directions. Thus, the Akcelik 248
delay model which aims movement-related design was used for optimizing the signal timings 249
![Page 11: Design of an Efficient Vehicle-Actuated Signal Control ...scientiairanica.sharif.edu/article_22418_3cb77f0f300eb1046401b00b… · 1 1 Design of an Efficient Vehicle-Actuated Signal](https://reader035.vdocuments.mx/reader035/viewer/2022070221/61357ca30ad5d206764768b1/html5/thumbnails/11.jpg)
11
in this study. According to the Akcelik method, the approximate value of total delay for a 250
movement can be calculated by using Eq. (2) [42]: 251
( )
( )
21
2 1o
qC uD N x
y
−= +
− (2) 252
Where; 253
D : Total Delay (sec), 254
q : Flow (veh/sec), 255
C : Cycle time (sec), 256
qC : Average number of arrivals in vehicles per cycle, 257
u : Green time ratio ( g C ), 258
y : Flow ratio ( q s ), 259
oN : Average overflow queue (veh) 260
x : Degree of saturation ( q Q ) (veh). 261
At a signalized intersection approach, to predict the overflow queues for both undersaturated 262
( 1x ) and oversaturated ( 1x ) conditions, the following Eq. (3) which was developed by 263
Akcelik can be used: 264
( )212
4
0
f o
o
O f
o
QT x xz z , x x
N QT
, x x
− + +
=
(3) 265
Where; 266
oN : Average overflow queue (veh), 267
Q : Capacity in vehicle in per hour (veh/hr), 268
fT : Flow period, 269
fQT : Maximum number of vehicles which can be discharged during the interval fT , 270
![Page 12: Design of an Efficient Vehicle-Actuated Signal Control ...scientiairanica.sharif.edu/article_22418_3cb77f0f300eb1046401b00b… · 1 1 Design of an Efficient Vehicle-Actuated Signal](https://reader035.vdocuments.mx/reader035/viewer/2022070221/61357ca30ad5d206764768b1/html5/thumbnails/12.jpg)
12
x : Degree of saturation ( q Q ), 271
z : 1x− , 272
ox : The degree of saturation below which the average overflow queue is approximately 273
zero. 274
The degree of saturation below which the average overflow queue is approximately zero ( ox ) 275
can be calculated by Eq. (4): 276
0 67600
o
sgx .= + (4) 277
Where; 278
s : Saturation flow (veh/sec), 279
g : Effective green time (sec). 280
Average vehicle delay can be calculated by Eq. (5): 281
Dd
q= (5) 282
Where; 283
D : Total Delay (sec), 284
q : Flow (veh). 285
With the analysis studies, the determination of the effectiveness of VAM4 which was created 286
in VISVAP was aimed. Therefore, firstly, it was assumed that the modeled signalized 287
intersection was managed with an optimum fixed-time signal control system and it was operated 288
with four phases. The signal timings were optimized by minimizing of average vehicle delay at 289
the intersection. The objective function ( )f x , the decision variables and the set of constraints 290
for the signal timing optimization problem are shown in Table 3. 291
As can be seen from Table 3, for the modeled intersection, the optimization problem consists 292
of 1 objective function, 4 decision variables and 18 constraints. When the previous studies were 293
![Page 13: Design of an Efficient Vehicle-Actuated Signal Control ...scientiairanica.sharif.edu/article_22418_3cb77f0f300eb1046401b00b… · 1 1 Design of an Efficient Vehicle-Actuated Signal](https://reader035.vdocuments.mx/reader035/viewer/2022070221/61357ca30ad5d206764768b1/html5/thumbnails/13.jpg)
13
investigated carefully, it was seen that green signal timings for each phase should be within a 294
certain range [43]. Thus, green signal timings were constrained between 7 seconds and 45 295
seconds considering the previous studies in the literature [43, 44]. In addition to this, the degree 296
of saturation for each lane was constrained with a maximum of 1.2. Differential Evolution (DE) 297
algorithm which is a powerful and simple meta-heuristic optimization algorithm is preferred to 298
solve the signal timing optimization problem [45-49]. A signal timing optimization program 299
was developed in MATLAB for this purpose. For created traffic scenarios, obtained signal 300
timings and average vehicle delays are given in Table 4 in detail. 301
Because all of the scenarios which are created in the scope of the study are analyzed in VISSIM, 302
Akcelik average vehicle delay results and average vehicle delay results obtained from VISSIM 303
should be similar to each other for the same scenarios. Rational and suitable evaluation cannot 304
be possible, otherwise. Due to differences between the results, firstly, an evaluation area was 305
created considering the points which are 100 meters away from all approaches stop lines. Then, 306
driving behaviors and the safety factors in VISSIM software were revised considering the 307
results which are obtained from Akcelik delay equation. Thus, average vehicle delay results for 308
each scenario are likened to each other. For created scenarios, the comparison of the average 309
vehicle delay values is shown in Table 5. 310
As can be seen from Table 5, for created scenarios, the differences are below about 5% in 311
generally. The accuracy of the simulation is determined by using the Sum of Squared Error 312
value [50]. This value is calculated as 33.56. These results show that the VISSIM can provide 313
similar average delay values with the Akcelik delay equation and can be used for analysis 314
studies. The average vehicle delay values which are obtained from VISSIM and by using the 315
Akcelik delay equation are shown in Fig. 7(a). The percentage differences between Akcelik and 316
VISSIM are also shown in Fig. 7(b). 317
![Page 14: Design of an Efficient Vehicle-Actuated Signal Control ...scientiairanica.sharif.edu/article_22418_3cb77f0f300eb1046401b00b… · 1 1 Design of an Efficient Vehicle-Actuated Signal](https://reader035.vdocuments.mx/reader035/viewer/2022070221/61357ca30ad5d206764768b1/html5/thumbnails/14.jpg)
14
After the completion of the revision process in the VISSIM simulation program, to determine 318
the effectiveness of VAM4, created vehicle actuated signal management control algorithm is 319
tested on all traffic scenarios. At this stage, firstly, the most proper placements of detectors for 320
each scenario were determined with the trial and error method. For Scenario 5 and Scenario 17, 321
average vehicle delay values which were obtained depending on the detector placement are as 322
shown in Fig. 8. 323
As seen in Fig. 8, a comprehensive study was carried out to determine the most proper 324
placement of the detectors. Because the distance between detectors and stop lines is varied from 325
15 meters to 45 meters, detectors must be placed within this range at all intersection approaches. 326
At this stage, created scenarios were analyzed considering different distances between detectors 327
and stop lines. At the end of the analyses, the distance which corresponds to the minimum 328
average vehicle delay value was considered as the most proper distance between detectors and 329
stop lines. Near optimum detector placements (distances) which are obtained from the analysis 330
for all created scenarios are presented in Table 6. 331
As can be seen from Table 6, when the total traffic volume at the intersection increases, the 332
distance between detectors and stop line also shows an increase. While the total traffic demand 333
at the intersection is low or moderate, near optimum distances between detectors and stop line 334
vary in the range of 15 meters and 25 meters in generally. Besides, while the total traffic demand 335
at the intersection is high or very high, near optimum distances between detectors and stop line 336
vary in the range of between 30 meters and 45 meters. 337
4. Results 338
After the determination of the most proper placements of detectors for all created scenarios, 339
two different types of traffic management approaches were evaluated separately considering 340
whether the fluctuations in traffic demands exist or not for each scenario. These four different 341
cases which are taken into account for the analysis can be summarized as follows: 342
![Page 15: Design of an Efficient Vehicle-Actuated Signal Control ...scientiairanica.sharif.edu/article_22418_3cb77f0f300eb1046401b00b… · 1 1 Design of an Efficient Vehicle-Actuated Signal](https://reader035.vdocuments.mx/reader035/viewer/2022070221/61357ca30ad5d206764768b1/html5/thumbnails/15.jpg)
15
• Optimum Fixed-Time Management / Fluctuations do not exist in traffic demands (OFTM 343
/ Not-Fluctuated): In this case, fluctuations do not exist in traffic demands. Movement-344
based traffic volumes and optimum signal timings are transferred to VISSIM. At the end 345
of the analysis, average vehicle delay values for each scenario were obtained. Obtained 346
results can be seen in Table 5. 347
• Vehicle-Actuated Management / Fluctuations do not exist in traffic demands (VAM4 / 348
Not-Fluctuated): In the second case, fluctuations do not exist in traffic demands. All 349
scenarios were analyzed considering the VAM4 model which is created in VISVAP. 350
• Optimum Fixed-Time Management / Fluctuations exist in traffic demands (OFTM / 351
Fluctuated): In the third case, fluctuations exist in traffic demands for each scenario (in 352
Fig. 6). The only difference of this case from the first case is existing fluctuations in traffic 353
demands. Movement-based traffic volumes and total traffic volumes are the same for the 354
first case and third case. In this case, the investigation of the success of optimum fixed-355
time management (OFTM) was aimed for fluctuated traffic demands. 356
• Vehicle-Actuated Management / Fluctuations exist in traffic demands (VAM4 / 357
Fluctuated): In the fourth case, fluctuations exist in traffic demands for each scenario (in 358
Fig. 6). The only difference of this case from the second case is existing fluctuations in 359
demands. Movement-based traffic volumes and total traffic volumes are the same for the 360
second case and fourth case. In this case, the investigation of the success of VAM4 was 361
aimed for fluctuated traffic demands. 362
At the end of the analysis which is made considering four different cases described above, 363
average vehicle delay-fuel consumption and exhaust emission (CO+NOx) values were obtained 364
for each scenario. Obtained average vehicle delay, fuel consumption and exhaust emission 365
values are shown in Fig. 9(a), Fig. 9(b) and Fig. 9(c), respectively. 366
![Page 16: Design of an Efficient Vehicle-Actuated Signal Control ...scientiairanica.sharif.edu/article_22418_3cb77f0f300eb1046401b00b… · 1 1 Design of an Efficient Vehicle-Actuated Signal](https://reader035.vdocuments.mx/reader035/viewer/2022070221/61357ca30ad5d206764768b1/html5/thumbnails/16.jpg)
16
As seen in Fig. 9, especially in case of fluctuated traffic demands, the effectiveness of the 367
optimum fixed-time signal management system decreases dramatically. Besides, the average 368
vehicle delays, fuel consumptions and exhaust emissions can be significantly reduced by 369
implementing vehicle-actuated signal control at the intersection. For Scenario 3, in the time 370
interval when the fluctuations occur in traffic demands, sample intersection views which are 371
obtained from VISSIM in case of implementing the optimum fixed-time management or/and 372
vehicle actuated management are presented in Fig. 10. 373
In Fig. 10, for Scenario 3, in the time interval when the fluctuations occur in the traffic demands, 374
sample intersection views in case of implementing the different types of management 375
approaches are shown. As shown in Fig. 6, the fluctuations in the traffic demands for Scenario 376
3 occur in between the 12th and 36th minutes. Therefore, simulation views are taken in the 25th 377
minutes (1475 sec.) of the simulation. As can be seen from Fig. 10, in case of implementing of 378
optimum-fixed time traffic management at modeled signalized intersection, excessive queuing 379
(queues) are seen at the intersection approaches where fluctuations in traffic demands occur 380
(Fig. 10 (a)). When Fig. 10(b) is examined carefully, in case of implementing vehicle-actuated 381
management instead of optimum fixed-time management at the modeled intersection, queuing 382
(queues) at intersection approaches where fluctuations in traffic demands occur can be 383
significantly reduced. Thus, the performance of the intersection can also be improved in this 384
way. 385
When the fluctuations do not exist in traffic demands, although vehicle-actuated signal 386
management systems have more advantages than optimum fixed-time signal management 387
systems, obtained success rates become low. For the sample scenario (Scenario 3) which is 388
shown in Fig. 10, sample intersection views in the same time intervals in case of implementing 389
the optimum fixed-time management and vehicle-actuated management are presented in Fig. 390
11. 391
![Page 17: Design of an Efficient Vehicle-Actuated Signal Control ...scientiairanica.sharif.edu/article_22418_3cb77f0f300eb1046401b00b… · 1 1 Design of an Efficient Vehicle-Actuated Signal](https://reader035.vdocuments.mx/reader035/viewer/2022070221/61357ca30ad5d206764768b1/html5/thumbnails/17.jpg)
17
In Fig. 11, for Scenario 3, when the fluctuations do not exist in the traffic demands, sample 392
intersection views in the same time interval in case of implementing the different types of 393
management approaches are shown. As can be seen from Fig. 11, when the fluctuations do not 394
exist in traffic demands, excessive queuing is not seen at the intersection approaches in case of 395
implementing of both optimum fixed-time management and vehicle-actuated management. But, 396
more queuing occurs in case of implementing of optimum fixed-time management (in Fig. 397
11(a)) than in case of implementing of vehicle-actuated management (in Fig. 11(b)). When Fig. 398
11 is investigated carefully, the queuing at intersection approaches can be reduced in case of 399
implementing of vehicle-actuated management. As a result, it was concluded that simulation 400
views are compatible with obtained performance results. The simulation views support and 401
verify the obtained results. 402
For created traffic scenarios, reduction rates (success rates) were evaluated separately 403
considering whether fluctuations in traffic demands exist or not. In case of implementing of 404
vehicle-actuated management instead of optimum fixed-time management, obtained reduction 405
rates (success rates) for each performance criteria are presented in Fig. 12. 406
When Fig. 12 is examined carefully, the obtained results can be summarized as follows: 407
• For the fluctuated traffic demands, when the total traffic demand at the intersection is 408
low, moderate or high, average vehicle delay can be reduced between about 20% and 409
60%, fuel consumption and exhaust emission can be reduced between about 15% and 410
40%. 411
• For the fluctuated traffic demands, when the total traffic demand at the intersection is 412
very high, the reduction rates for average vehicle delay decrease down to 15%, reduction 413
rates for fuel consumption and exhaust emission decrease down to 10%. These results 414
show that the vehicle-actuated management systems are not efficient systems for 415
specified traffic demand conditions. 416
![Page 18: Design of an Efficient Vehicle-Actuated Signal Control ...scientiairanica.sharif.edu/article_22418_3cb77f0f300eb1046401b00b… · 1 1 Design of an Efficient Vehicle-Actuated Signal](https://reader035.vdocuments.mx/reader035/viewer/2022070221/61357ca30ad5d206764768b1/html5/thumbnails/18.jpg)
18
• For not-fluctuated traffic demands, when the total traffic demand at the intersection is 417
low, moderate or high, average vehicle delay can be reduced between about 5% and 20%, 418
fuel consumption and exhaust emission can be reduced between about 3% and 9%. 419
• For the not-fluctuated traffic demands, when the total traffic demand at the intersection 420
is very high, it can be seen that the reduction rates for average vehicle delay increase up 421
to 30%, reduction rates for fuel consumption and exhaust emission increase up to 18%. 422
These results show that the vehicle-actuated management systems are very effective 423
systems for specified traffic demand conditions. 424
When the cases of whether the fluctuations in traffic demands exist or not are investigated 425
separately, the obtained results are presented in Table 7 in detail. 426
As seen in Table 7, for the created scenarios; 427
• For not-fluctuated traffic demands, in case of implementing of VAM4 instead of OFTM, 428
the obtained reduction rates for average vehicle delay are about 4.5% of minimum, 29.2% 429
of maximum and 14.2% of average. For the fuel consumption and exhaust emission, these 430
rates are about 2.8% of minimum, 17.7% of maximum and 7.3% of average. 431
• For fluctuated traffic demands, in case of implementing of VAM4 instead of OFTM, the 432
obtained reduction rates for average vehicle delay are about 15.6% of minimum, 58.6% of 433
maximum and 32.5% of average. For the fuel consumption and exhaust emission, these 434
rates are about 9.6% of minimum, 38.4% of maximum and 21.8% of average. 435
As can be seen from the obtained results, developed VAM4 is more effective in case of 436
fluctuated traffic demands. For fluctuated traffic demands, reduction rates for all performance 437
criteria increase about three times compared to not-fluctuated traffic demands in case of 438
implementing of VAM4 instead of OFTM. 439
5. Discussion and Conclusion 440
![Page 19: Design of an Efficient Vehicle-Actuated Signal Control ...scientiairanica.sharif.edu/article_22418_3cb77f0f300eb1046401b00b… · 1 1 Design of an Efficient Vehicle-Actuated Signal](https://reader035.vdocuments.mx/reader035/viewer/2022070221/61357ca30ad5d206764768b1/html5/thumbnails/19.jpg)
19
In this study, a VAM4 that can adapt to fluctuations in traffic demands was developed. The 441
effectiveness of the developed system was tested considering whether the fluctuations in traffic 442
demands exist or not. Obtained results from this study can be summarized as follows: 443
• When the fluctuations in traffic demands did not exist, lower average vehicle delay, fuel 444
consumption and exhaust emission values were obtained with VAM4 compared to 445
OFTM. When Fig. 12 is examined carefully, especially in case of traffic demand is very 446
high at the intersection, reduction rates for all performance criteria increase. One of the 447
main reasons for this situation is that the constraint of maximum green time of 45 seconds 448
for OFTM is lower than the constraint of maximum green time of 60 seconds for VAM4. 449
The increase of maximum green time length for VAM4 enabled to lower waste of time 450
for vehicles at intersection approaches where the traffic demand is very high. Thus, 451
average vehicle delays, fuel consumption and exhaust emissions were significantly 452
reduced. 453
• When the fluctuations in traffic demands existed, the effectiveness of OFTM decreased 454
and average vehicle delay, fuel consumptions and exhaust emissions significantly 455
increased. Average vehicle delay, fuel consumption and exhaust emissions were 456
significantly reduced with the implementation of VAM4, especially in case of traffic 457
demand at the intersection was low, moderate and high. In case of traffic demand at the 458
intersection was very high, reduction rates decreased. This situation has arisen from the 459
increase of the usage of maximum green time by different intersection approaches. It was 460
determined that the density of the usage of maximum green time was an important factor 461
that encourages the increase of delay, fuel consumption and exhaust emission. 462
• For created scenarios, when the fluctuations in traffic demands did not exist, in case of 463
implementation of VAM4 instead of OFTM, the average vehicle delay was reduced about 464
the average of 14.2%. Besides, fuel consumption and exhaust emissions were reduced by 465
![Page 20: Design of an Efficient Vehicle-Actuated Signal Control ...scientiairanica.sharif.edu/article_22418_3cb77f0f300eb1046401b00b… · 1 1 Design of an Efficient Vehicle-Actuated Signal](https://reader035.vdocuments.mx/reader035/viewer/2022070221/61357ca30ad5d206764768b1/html5/thumbnails/20.jpg)
20
about an average of 7.3%. When the fluctuations in traffic demand existed, these rates 466
increased 2.5-3 times. 467
• When the fluctuations in traffic demands did not exist, in case of implementation of 468
VAM4 instead of OFTM, the level of service of only five scenarios was improved. When 469
the fluctuations in traffic demands existed, the level of service of fourteen scenarios was 470
improved. 471
All obtained results show that VAM4 can effectively adapt to fluctuations in traffic demands. 472
In addition to this, it is thought that the effectiveness and the performance of VAM4 can be 473
more improved with the different computational approaches. 474
Acknowledgments 475
The authors of this article would like to thank to PTV (Planung Transport Verkehr AG) for 476
providing of VISSIM simulation software. 477
Funding 478
The authors of this article received no financial support for the research, authorship, and 479
publication of this article. 480
Availability of Data and Materials 481
The data used to support the findings of this study are available from the corresponding author 482
upon request. 483
6. References 484
[1] Li, J-Q., Wu, G. and Zou, N. “Investigation of The Impacts of Signal Timing on Vehicle 485
Emissions at an Isolated Intersection”, Transp. Res. D Transp. Environ., 16 (2011), pp. 409-486
414 (2011). 487
[2] Asadi, H., Tavakkoli Moghaddam, R. and Shahsavari Pour, N. et al., “A New 488
Nondominated Sorting Genetic Algorithm based to The Regression Line for Fuzzy Traffic 489
Signal Optimization Problem”, Sci. Iran., 25 (3), pp. 1712-1723 (2018). 490
![Page 21: Design of an Efficient Vehicle-Actuated Signal Control ...scientiairanica.sharif.edu/article_22418_3cb77f0f300eb1046401b00b… · 1 1 Design of an Efficient Vehicle-Actuated Signal](https://reader035.vdocuments.mx/reader035/viewer/2022070221/61357ca30ad5d206764768b1/html5/thumbnails/21.jpg)
21
[3] Yang, L., Zhang, L. and Stettler, M. E. J. et al., “Supporting An Integrated Transportation 491
Infrastructure and Public Space Design: A Coupled Simulation Method for Evaluating 492
Traffic Pollution and Microclimate”, Sustain. Cities Soc., 52 (2020), pp. 101796 (2020). 493
[4] Barzegar, S., Davoudpour, M. and Meybodi, M. R. et al., “Formalized Learning Automata 494
with Adaptive Fuzzy Coloured Petri net; An Application Specific to Managing Traffic 495
Signals”, Sci. Iran., 18 (3), pp. 554-565 (2011). 496
[5] Samadi, S., Rad, A. P. and Kazemi, F. M. et al., “Performance Evaluation of Intelligent 497
Adaptive Traffic Control Systems: A Case Study”, J. Transp. Technol., 2, pp. 248-259 498
(2012). 499
[6] McKenny, D. and White, T. “Distributed and Adaptive Traffic Signal Control Within a 500
Realistic Traffic Simulation”, Eng. Appl. Artif. Intell., 26, pp. 574-583 (2012). 501
[7] Vilarinho, C. and Tavares, J. P. “Real-Time Traffic Signal Settings at an Isolated Signal 502
Control Intersection”, Procedia Soc. Behav. Sci., 3, pp. 1021-1030 (2014). 503
[8] Dabiri, S. and Abbas, M. “Arterial Traffic Signal Optimization using Particle Swarm 504
Optimization in an Integrated VISSIM-MATLAB Simulation Environment”, 2016 IEEE 505
19th International Conference on Intelligent Transportation Systems, Rio de Janerio, Brazil, 506
pp. 766-771 (2016). 507
[9] Abdelghaffar, H. M., Yang, H. and Rakha, H. A. “Isolated Traffic Signal Control using 508
Nash Bargaining Optimization”, Glob. J. Res. Eng. B, 16 (1), pp. 1-11 (2016). 509
[10] Li, Z., Shahidehpour, M. and Bahramirad, S. et al., “Optimizing Traffic Signal Settings 510
in Smart Cities”, IEEE T. Smart Grid, 8 (5), pp. 2382-2393 (2017). 511
[11] Hashim, I., Ragab, M. and Asar, G. “Optimization of Vehicle Delay and Exhaust 512
Emissions at Signalized Intersections”, Adv. Transp. Stud. A, 46 (2018), 5-18 (2018). 513
![Page 22: Design of an Efficient Vehicle-Actuated Signal Control ...scientiairanica.sharif.edu/article_22418_3cb77f0f300eb1046401b00b… · 1 1 Design of an Efficient Vehicle-Actuated Signal](https://reader035.vdocuments.mx/reader035/viewer/2022070221/61357ca30ad5d206764768b1/html5/thumbnails/22.jpg)
22
[12] Yao, Z., Jiang, Y. and Zhao, B. et al., “A Dynamic Optimization Method for Adaptive 514
Signal Control in a Connected Vehicle Environment”, J. Intell. Transp. Syst., 24 (2), pp. 515
184-200 (2019). 516
[13] Grether, D., Bischoff, J. and Nagel, K. “Traffic Actuated Signal Control: Simulation of 517
The User Benefits in a Big Event Real-World Scenario”, 2nd International Conference on 518
Models and Technologies for Intelligent Transportation Systems, pp. 1-4 (2011). 519
[14] Bullen, A. G. R. “Effects of Actuated Signal Settings and Detector Placement on 520
Vehicle Delay”, Transp. Res. Rec., 1244, pp. 32-38 (1989). 521
[15] Mathew, T. V. Signalized Intersections Delay Models, Lecture Notes in Transportation 522
System Engineering (2017). 523
[16] Villiers, C., Nguyen, L. D. and Zalewski, J. “Evaluation of Traffic Management 524
Strategies for Special Events using Probe Data”, TRIP, 2 (2019), pp. 100052 (2019). 525
[17] Van Katwijk, R. T. Multi-Agent Look-Ahead Traffic-Adaptive Control, PhD Thesis, 526
Technische Universiteit Delft, Delft, (2008). 527
[18] Gundogan, F. Simplified Traffic Responsive Signal Control Method for Developing 528
Large Cities, PhD Thesis, Graz University of Technology, Graz, (2012). 529
[19] Webster, F. V. Traffic Signal Settings Report (1958). 530
[20] Mohajerpoor, R., Saberi, M. and Ramezani, M. “Analytical Derivation of the Optimal 531
Traffic Signal Timing: Minimizing Delay Variability and Spillback Probability for 532
Undersaturated Intersections”, Transp. Res. B Meth., 119 (2019), pp. 45-68 (2019). 533
[21] Lian, F., Chen, B. and Zhang, K. et al., “Adaptive Traffic Signal Control Algorithms 534
based on Probe Vehicle Data”, J. Intell. Transp. Syst., 25 (2021), pp. 41-57 (2021). 535
[22] Akcelik, R. “Estimation of Green Times and Cycle Time for Vehicle Actuated Signals”, 536
Transp. Res. Rec., 1457, 63-72 (1994). 537
![Page 23: Design of an Efficient Vehicle-Actuated Signal Control ...scientiairanica.sharif.edu/article_22418_3cb77f0f300eb1046401b00b… · 1 1 Design of an Efficient Vehicle-Actuated Signal](https://reader035.vdocuments.mx/reader035/viewer/2022070221/61357ca30ad5d206764768b1/html5/thumbnails/23.jpg)
23
[23] Mirchandani, P. and Head, L. “A Real-Time Traffic Signal Control System: 538
Architecture, Algorithms and Analysis”, Transp. Res. C Emerg. Technol., 9 (2001), pp. 539
415-432 (2001). 540
[24] Angulo, E., Romero, F. P. and Garcia, R. et al., “An Adaptive Approach to Enhanced 541
Traffic Signal Optimization by using Soft-Computing Techniques”, Expert Syst. Appl., 38, 542
2235-2247 (2011). 543
[25] Yu, D., Tian, X. and Xing, X. et al., “Signal Timing Optimization Based on Fuzzy 544
Compromise Programming for Isolated Signalized Intersection”, Math. Probl. Eng., 2016, 545
Article ID: 1682394, pp. 1-12 (2016). 546
[26] Hao, R., Wang, L. and Ma, V. et al., “Estimating Signal Timing of Actuated Signal 547
Control using Pattern Recognition under Connected Vehicle Environment”, Promet-548
Zagreb, 33 (1), pp. 153-163 (2021). 549
[27] Li, H. and Prevedouros, P. D. “Traffic Adaptive Control for Oversaturated Isolated 550
Intersections: Model Development and Simulation Testing”, J. Transp. Eng., 130 (5), pp. 551
594-601 (2004). 552
[28] Viti, F. and van Zuylen, H. J. “A Probabilistic Model for Traffic at Actuated Control 553
Signals”, Transp. Res. C Emerg. Technol., 18 (2010), pp. 299-310 (2010). 554
[29] Trabia, M. B., Kaseko, M. S. and Ande, M. “A Two-Stage Fuzzy Logic Controller for 555
Traffic Signals”, Transp. Res. C Emerg. Technol., 7 (1999), pp. 353 – 367 (1999). 556
[30] Kim, J. T. and Courage, K. G. “Evaluation and Design of Maximum Green Time 557
Settings for Traffic-Actuated Control”, Transp. Res. Rec., 3606, pp. 246-255 (2003). 558
[31] Guo, W., Yu, Z. and He, Z. et al., “Traffic-Actuated Signal Control based on Dynamic 559
Optimal Maximum Green Time”, International Conference on Transportation Engineering 560
2007 (ICTE 2007), pp. 582-587 (2007). 561
![Page 24: Design of an Efficient Vehicle-Actuated Signal Control ...scientiairanica.sharif.edu/article_22418_3cb77f0f300eb1046401b00b… · 1 1 Design of an Efficient Vehicle-Actuated Signal](https://reader035.vdocuments.mx/reader035/viewer/2022070221/61357ca30ad5d206764768b1/html5/thumbnails/24.jpg)
24
[32] Zheng, X., Recker, W. and Chu, L. “Optimization of Control Parameters for Adaptive 562
Traffic-Actuated Signal Control”, J. Intell. Transp. Syst., 14 (2), pp. 95-108 (2010). 563
[33] Jiang, X., Qiu, Y. and Ruan, S. “An Approach to Optimize the Settings of Actuated 564
Signals”, J. Mod. Transp., 19 (1), pp. 68-74 (2011). 565
[34] Swaminathan, N., Rathinawel, N. and Duraisamy, S. et al., “Design of Vehicle Actuated 566
Signal Using Simulation”, Gradevinar, 7 (2014), pp. 635-641 (2014). 567
[35] Guo, Y. and Ma, J. “An Improved Actuated Signal Control of Intersection based on 568
VISVAP”, 6th International Conference on Sensor Network and Computer Engineering, pp. 569
123-128 (2016). 570
[36] Ribeiro, I. M. and Simoes, M. L. O. “The Fully Actuated Traffic Control Problem 571
Solved by Global Optimization and Complementarity”, Eng. Optim., 48 (2), pp. 199-212 572
(2016). 573
[37] Lee, S. H. and Han, D. “Safety Impacts of The Actuated Signal Control at Urban 574
Intersections”, Promet-Zagreb, 28 (1), pp. 31-39 (2016). 575
[38] Wang, X.B, Yin, K. and Liu, H. “Vehicle Actuated Signal Performance Under General 576
Traffic at an Isolated Intersection”, Transp. Res. C Emerg. Technol., 95 (2018), pp. 582-577
598 (2018). 578
[39] Promraksa, T., Satiennam, T. and Satiennam, W. “Vehicle Actuated Signal Control for 579
Low Carbon Society”, Int. J. GEOMATE, 16 (55), pp. 86-91 (2019). 580
[40] Highway Capacity Manual (2010). National Research Council, Washington, D. C., 581
(2010). 582
[41] Dogan, E. Isolated Signalized Intersection Control by Optimized Fuzzy Logic Method, 583
PhD Thesis, Kirikkale University, Institute of Natural and Applied Sciences, Kırıkkale, 584
(2014). 585
![Page 25: Design of an Efficient Vehicle-Actuated Signal Control ...scientiairanica.sharif.edu/article_22418_3cb77f0f300eb1046401b00b… · 1 1 Design of an Efficient Vehicle-Actuated Signal](https://reader035.vdocuments.mx/reader035/viewer/2022070221/61357ca30ad5d206764768b1/html5/thumbnails/25.jpg)
25
[42] Akcelik, R. Traffic Signals: Capacity and Timing Analysis, Australian Road Research 586
Board, Research Report ARR No: 123 (1998). 587
[43] Murat, Y. S. and Kikuchi, S. “The Fuzzy Optimization Approach: A Comparison with 588
the Classical Optimization Approach Using the Problem of Timing a Traffic Signal”, 589
Transp. Res. Rec., 2024, pp. 82-91 (2007). 590
[44] Ceylan, H. and Bell, M. G. H. “Traffic Signal Timing Optimisation based on Genetic 591
Algorithm Approach including Drivers’ Routing”, Transp. Res. B Meth., 38 (4), pp. 329-592
342 (2004). 593
[45] Keskinturk, T. “Differential Evolution Algorithm”, J Ins. Nat. App. Sci. of Istanbul 594
Ticaret University, 9 (2006), pp. 85-99 (2007). 595
[46] Baskan, O. and Ceylan, H. “Differential Evolution Algorithm based Solution 596
Approaches for Solving Transportation Network Design Problems”, Pamukkale University 597
J. Eng. Sci., 20 (9), pp. 324-331 (2014). 598
[47] Dogan, E. and Akgungor, A. P. “Optimizing a Fuzzy Logic Traffic Signal Controller 599
via the Differential Evolution Algorithm Under Different Traffic Scenarios”, Simulation, 600
92 (11), pp. 1013-1023 (2016). 601
[48] Korkmaz, E. and Akgungor, A. P. “Delay Estimation Models for Signalized 602
Intersections Using Differential Evolution Algorithm”, Int. J. Eng. Res., 5 (3), pp. 16-29 603
(2017). 604
[49] Baskan, O. and Ozan, C. “Determining Optimum Configuration of One-Way and Two-605
Way Streets Using Shortest Path Travel Costs Based on Results of Traffic Assignment”, 606
Pamukkale University J. Eng. Sci., 24 (6), pp. 1087-1092 (2018). 607
[50] Yu, L., Yu, L. and Chen, X. et al., “Calibration of VISSIM for Bus Rapid Transit 608
Systems in Beijing using GPS Data”, J. Public Trans., 2006 BRT Special Edition, pp. 239-609
257 (2006). 610
![Page 26: Design of an Efficient Vehicle-Actuated Signal Control ...scientiairanica.sharif.edu/article_22418_3cb77f0f300eb1046401b00b… · 1 1 Design of an Efficient Vehicle-Actuated Signal](https://reader035.vdocuments.mx/reader035/viewer/2022070221/61357ca30ad5d206764768b1/html5/thumbnails/26.jpg)
26
Figure and Table Captions 611
List of Figures 612
Fig. 1 The relation between average vehicle delay and the placement of detectors 613
Fig. 2 Vehicle-actuated management application at a four-leg signalized intersection in VISSIM 614
Fig. 3 The flow chart of vehicle-actuated control logic created with VAP 615
Fig. 4 The change rates of movement direction-based traffic volumes for created scenarios 616
Fig. 5 Hourly total traffic volumes for all traffic scenarios 617
Fig. 6 Time intervals and quantities of fluctuations in traffic demands for each traffic scenario 618
Fig. 7 For created scenarios (a) the comparison of average vehicle delay values and (b) the 619
percentage (%) differences of average vehicle delay values. 620
Fig. 8 The effect of detector placements on average vehicle delay (a) for Scenario 5 and (b) 621
Scenario 17. 622
Fig. 9 At the end of the analysis, for created scenarios (a) obtained average vehicle delay results; 623
(b) obtained fuel consumption results; (c) obtained exhaust emission results. 624
Fig. 10 For the Scenario 3, in time interval when the fluctuations occur in traffic demands, 625
sample intersection view in case of implementing (a) the optimum-fixed time management 626
(OFTM) (b) vehicle-actuated management (VAM4) 627
Fig. 11 For Scenario 3, when the fluctuations do not exist in traffic demands, sample 628
intersection view in time same time interval in case of implementing (a) the optimum fixed-629
time management (OFTM), (b) vehicle-actuated management (VAM4) 630
Fig. 12 Obtained reduction rates in case of implementing of vehicle-actuated management 631
instead of optimum fixed-time management for each performance criteria 632
List of Tables 633
Table 1 Minimum green times for different detector locations 634
Table 2 Reference demand matrix created for analysis studies 635
![Page 27: Design of an Efficient Vehicle-Actuated Signal Control ...scientiairanica.sharif.edu/article_22418_3cb77f0f300eb1046401b00b… · 1 1 Design of an Efficient Vehicle-Actuated Signal](https://reader035.vdocuments.mx/reader035/viewer/2022070221/61357ca30ad5d206764768b1/html5/thumbnails/27.jpg)
27
Table 3 The objective function, the decision variables and the set of constraints for the signal 636
timing optimization problem 637
Table 4 Optimum signal timings and average vehicle delays for created traffic scenarios 638
Table 5 Scenario-based comparison of average vehicle delay values (sec/veh) 639
Table 6 Optimum detector placements for created scenarios 640
Table 7 For the cases of fluctuated and not-fluctuated traffic demands, obtained reduction rates 641
(success rates) 642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
![Page 28: Design of an Efficient Vehicle-Actuated Signal Control ...scientiairanica.sharif.edu/article_22418_3cb77f0f300eb1046401b00b… · 1 1 Design of an Efficient Vehicle-Actuated Signal](https://reader035.vdocuments.mx/reader035/viewer/2022070221/61357ca30ad5d206764768b1/html5/thumbnails/28.jpg)
28
661
662
663
664
665
666
Fig. 1 The relation between average vehicle delay and the placement of detectors 667
668
669
670
671
672
673
674
675
676
677
678
679
Fig. 2 Vehicle-actuated management application at a four-leg signalized intersection in 680
VISSIM 681
682
683
684
685
Near minimum delay Near optimum distance
Av
era
ge
Del
ay
Distance between Detectors and Stop Line
North East
West
South
Phase I
Phase II
Phase III
Phase IV
Detector No 1 – 2 – 3 – 4
Detector No 5 – 6 – 7
Detector No 8 – 9 – 10 - 11
Detector No 12 – 13 – 14
VAM4
![Page 29: Design of an Efficient Vehicle-Actuated Signal Control ...scientiairanica.sharif.edu/article_22418_3cb77f0f300eb1046401b00b… · 1 1 Design of an Efficient Vehicle-Actuated Signal](https://reader035.vdocuments.mx/reader035/viewer/2022070221/61357ca30ad5d206764768b1/html5/thumbnails/29.jpg)
29
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
Fig. 3 The flow chart of vehicle-actuated control logic created with VAP 709
710
N
N
Y
Y
N
Y Y
N
N
N
Y Y
Y Y
Y Y
N
N
Y Y
N
Y
Y
N
N
N
N
N
N
Y Y
Y Y
Y Y
Y Y
N
Y Y
N
N
N
N
N
Y Y
N
Y Y
Y Y
Y Y
N
Y Y
N
N
N
N
N
N
Y Y
Y Y
Y Y
N
Y
Start (t=1)
Any interphase active?
Phase I active?
Phase II active?
Phase III active?
Intersection management period complete? (t=3600 sec.)
Obtain the average vehicle delay from VISSIM simulation program
Green Time
for Phase I
>=
Minimum
Green Time
for Phase I
Green Time
for Phase I
<
Maximum Green Time
for Phase I
Min. critical gap (1 or 2 or 3 or 4)
<=
⁓ 2 sec.
Occupancy (5 or 6 or 7)
>=
1 sec.
Occupancy (8 or 9 or 10 or 11)
>=
1 sec.
Occupancy (12 or 13 or 14)
>=
1 sec.
From Phase I
to Phase II
From Phase I
to Phase III
From Phase I
to Phase IV
Green Time
for Phase II
>=
Minimum
Green Time
for Phase II
Green Time
for Phase II
<
Maximum Green Time
for Phase II
Min. critical gap (5 or 6 or 7)
<=
⁓ 2 sec.
Occupancy (8 or 9 or 10 or 11)
>=
1 sec.
Occupancy (12 or 13 or 14)
>=
1 sec.
Occupancy (1 or 2 or 3 or 4)
>=
1 sec.
From Phase II
to Phase III
From Phase II
to Phase IV
From Phase II
to Phase I
Green Time
for Phase III
>=
Minimum
Green Time
for Phase III
Green Time
for Phase III
<
Maximum Green Time
for Phase III
Min. critical gap (8 or 9 or 10 or 11)
<=
⁓ 2 sec.
Occupancy (12 or 13 or 14)
>=
1 sec.
Occupancy (1 or 2 or 3 or 4)
>=
1 sec.
Occupancy (5 or 6 or 7)
>=
1 sec.
Phase IV active?
From Phase III
to Phase IV
From Phase III
to Phase I
From Phase III
to Phase II
Green Time
for Phase IV
>=
Minimum
Green Time
for Phase IV
Green Time
for Phase IV
<
Maximum Green Time
for Phase IV
Min. critical gap (12 or 13 or 14)
<=
⁓ 2 sec.
Occupancy (1 or 2 or 3 or 4)
>=
1 sec.
Occupancy (5 or 6 or 7)
>=
1 sec.
Occupancy (8 or 9 or 10 or 11)
>=
1 sec.
From Phase IV
to Phase I
From Phase IV
to Phase II
From Phase IV
to Phase III
t = t+1
![Page 30: Design of an Efficient Vehicle-Actuated Signal Control ...scientiairanica.sharif.edu/article_22418_3cb77f0f300eb1046401b00b… · 1 1 Design of an Efficient Vehicle-Actuated Signal](https://reader035.vdocuments.mx/reader035/viewer/2022070221/61357ca30ad5d206764768b1/html5/thumbnails/30.jpg)
30
711
Fig. 4 The change rates of movement direction-based traffic volumes for created scenarios 712
713
714
715
716
Fig. 5 Hourly total traffic volumes for all traffic scenarios 717
718
719
720
721
722
-50%
-40%
-30%
-20%
-10%
0%
10%
20%
30%
40%
50%
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Ch
an
ge
Ra
tes
Scenerio No
The Change Rates of Movement Direction Based Traffic Volumes
Turning Right Going Straight Turning Left
Reference Demand Matrix
0
500
1000
1500
2000
2500
3000
3500
4000
4500
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
To
tal
Tra
ffic
Vo
lum
e a
t
Inte
rsec
tio
n (
veh
/hr)
Scenario No
Hourly Total Traffic Volumes for all Traffic Scenarios
![Page 31: Design of an Efficient Vehicle-Actuated Signal Control ...scientiairanica.sharif.edu/article_22418_3cb77f0f300eb1046401b00b… · 1 1 Design of an Efficient Vehicle-Actuated Signal](https://reader035.vdocuments.mx/reader035/viewer/2022070221/61357ca30ad5d206764768b1/html5/thumbnails/31.jpg)
31
Scenario
No
Intersection
Approach
Analysis Period (60 minutes)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Scenario 1 West 44%
Scenario 2 West 33%
South 50%
Scenario 3 East 60%
Scenario 4 West 56%
East 56%
Scenario 5 North 33%
South 33%
Scenario 6
West 33% 28%
North 22% 22%
East 50%
Scenario 7
West 67%
East 42%
South 50%
Scenario 8
West 38%
North 38%
East 50%
South 38%
Scenario 9 West 50%
East 50%
Scenario 10 West 50%
South 50%
Scenario 11 North 67%
East 75%
Scenario 12 West 50%
East 50%
Scenario 13 West 42%
South 33% 33%
Scenario 14 North 50%
South 67%
Scenario 15 West 38%
Scenario 16 West 33%
Scenario 17 North 56%
South 38%
Scenario 18 North 67%
South 50%
Scenario 19 North 44%
South 33%
Scenario 20 East 50%
Fig. 6 Time intervals and quantities of fluctuations in traffic demands for each traffic scenario723
![Page 32: Design of an Efficient Vehicle-Actuated Signal Control ...scientiairanica.sharif.edu/article_22418_3cb77f0f300eb1046401b00b… · 1 1 Design of an Efficient Vehicle-Actuated Signal](https://reader035.vdocuments.mx/reader035/viewer/2022070221/61357ca30ad5d206764768b1/html5/thumbnails/32.jpg)
32
724
(a) (b) 725
Fig. 7 For created scenarios (a) the comparison of average vehicle delay values and (b) the 726
percentage (%) differences of average vehicle delay values 727
728
729
730
731
(a) (b) 732
Fig. 8 The effect of detector placements on average vehicle delay for (a) Scenario 5 and 733
(b) Scenario 17. 734
735
736
737
738
R² = 0.996
20
30
40
50
60
70
80
90
20 30 40 50 60 70 80 90
VIS
SIM
(se
c/veh
)
Akçelik (sec/veh)
VISSIM - Akçelik
The Comparison of Average Delay Values
-6%
-4%
-2%
0%
2%
4%
6%
1 2 3 4 5 6 7 8 9 1011121314151617181920
Per
cen
tag
e D
iffe
ren
ces
Scenario No
VISSIM - Akçelik
The Differences of Average Delay Values
21.76
20
22
24
26
28
30
32
15 20 25 30
Avera
ge V
eh
icle
Dela
y (
sec/v
eh
)
Distance between Detectors and Stop Line (m)
Scenario 5
63.762
66
70
74
78
35 40 45 50
Avera
ge V
eh
icle
Dela
y (
sec/v
eh
)
Distance between Detectors and Stop Line (m)
Scenario 17
![Page 33: Design of an Efficient Vehicle-Actuated Signal Control ...scientiairanica.sharif.edu/article_22418_3cb77f0f300eb1046401b00b… · 1 1 Design of an Efficient Vehicle-Actuated Signal](https://reader035.vdocuments.mx/reader035/viewer/2022070221/61357ca30ad5d206764768b1/html5/thumbnails/33.jpg)
33
739
740
(a) 741
742
743
744
745
746
747
748
0
15
30
45
60
1 2 3 4 5
Aver
ag
e V
ehic
le D
elay
(se
c/veh
)
Scenario
Average Vehicle Delay - Demand: Low
Not-Fluctuated (OFTM) Not-Fluctuated (VAM4)
Fluctuated (OFTM) Fluctuated (VAM4)
0
25
50
75
100
1 2 3 4 5
Aver
ag
e V
ehic
le D
elay
(se
c/veh
)
Scenario
Average Vehicle Delay - Demand: Moderate
Not-Fluctuated (OFTM) Not-Fluctuated (VAM4)
Fluctuated (OFTM) Fluctuated (VAM4)
0
25
50
75
100
1 2 3 4 5
Aver
ag
e V
ehic
le D
elay
(se
c/veh
)
Scenario
Average Vehicle Delay - Demand: High
Not-Fluctuated (OFTM) Not-Fluctuated (VAM4)
Fluctuated (OFTM) Fluctuated (VAM4)20
40
60
80
100
120
1 2 3 4 5
Aver
ag
e V
ehic
le D
elay
(se
c/veh
)
Scenario
Average Vehicle Delay - Demand: Very High
Not-Fluctuated (OFTM) Not-Fluctuated (VAM4)
Fluctuated (OFTM) Fluctuated (VAM4)
![Page 34: Design of an Efficient Vehicle-Actuated Signal Control ...scientiairanica.sharif.edu/article_22418_3cb77f0f300eb1046401b00b… · 1 1 Design of an Efficient Vehicle-Actuated Signal](https://reader035.vdocuments.mx/reader035/viewer/2022070221/61357ca30ad5d206764768b1/html5/thumbnails/34.jpg)
34
749
750
(b) 751
752
753
754
755
756
757
758
759
760
761
80
120
160
200
240
280
1 2 3 4 5
Fu
el C
on
sum
pti
on
(l/
hr)
Scenario
Fuel Consumption - Demand: Low
Not-Fluctuated (OFTM) Not-Fluctuated (VAM4)
Fluctuated (OFTM) Fluctuated (VAM4)100
150
200
250
300
350
400
450
1 2 3 4 5
Fu
el C
on
sum
pti
on
(l/
hr)
Scenario
Fuel Consumption - Demand: Moderate
Not-Fluctuated (OFTM) Not-Fluctuated (VAM4)
Fluctuated (OFTM) Fluctuated (VAM4)
150
200
250
300
350
400
450
500
1 2 3 4 5
Fu
el C
on
sum
pti
on
(l/
hr)
Scenario
Fuel Consumption - Demand: High
Not-Fluctuated (OFTM) Not-Fluctuated (VAM4)
Fluctuated (OFTM) Fluctuated (VAM4)250
300
350
400
450
500
550
1 2 3 4 5
Fu
el C
on
sum
pti
on
(l/
hr)
Scenario
Fuel Consumption - Demand: Very High
Not-Fluctuated (OFTM) Not-Fluctuated (VAM4)
Fluctuated (OFTM) Fluctuated (VAM4)
![Page 35: Design of an Efficient Vehicle-Actuated Signal Control ...scientiairanica.sharif.edu/article_22418_3cb77f0f300eb1046401b00b… · 1 1 Design of an Efficient Vehicle-Actuated Signal](https://reader035.vdocuments.mx/reader035/viewer/2022070221/61357ca30ad5d206764768b1/html5/thumbnails/35.jpg)
35
762
763
(c) 764
Fig. 9 At the end of the analysis, for created scenarios (a) obtained average vehicle delay 765
results; (b) obtained fuel consumption results; (c) obtained exhaust emission results 766
767
768
769
770
771
772
773
774
775
776
2000
3000
4000
5000
6000
1 2 3 4 5
Ex
hau
st E
mis
sion
(g
/hr)
Scenario
Exhaust Emission - Demand: Low
Not-Fluctuated (OFTM) Not-Fluctuated (VAM4)
Fluctuated (OFTM) Fluctuated (VAM4)
2000
4000
6000
8000
10000
1 2 3 4 5
Ex
hau
st E
mss
ion
(g
/hr)
Scenario
Exhaust Emission - Demand: Moderate
Not-Fluctuated (OFTM) Not-Fluctuated (VAM4)
Fluctuated (OFTM) Fluctuated (VAM4)
2000
4000
6000
8000
10000
12000
1 2 3 4 5
Ex
hau
st E
mis
sion
(g
/hr)
Scenario
Exhaust Emission - Demand: High
Not-Fluctuated (OFTM) Not-Fluctuated (VAM4)
Fluctuated (OFTM) Fluctuated (VAM4)5000
7000
9000
11000
13000
1 2 3 4 5
Ex
hau
st E
mis
sion
(g
/hr)
Scenario
Exhaust Emission - Demand: Very High
Not-Fluctuated (OFTM) Not-Fluctuated (VAM4)
Fluctuated (OFTM) Fluctuated (VAM4)
![Page 36: Design of an Efficient Vehicle-Actuated Signal Control ...scientiairanica.sharif.edu/article_22418_3cb77f0f300eb1046401b00b… · 1 1 Design of an Efficient Vehicle-Actuated Signal](https://reader035.vdocuments.mx/reader035/viewer/2022070221/61357ca30ad5d206764768b1/html5/thumbnails/36.jpg)
36
777
778
779
780
781
782
783
(a) (b) 784
Fig. 10 For the Scenario 3, in time interval when the fluctuations occur in traffic demands, 785
sample intersection view in case of implementing (a) the optimum-fixed time management 786
(OFTM) (b) vehicle-actuated management (VAM4). 787
788
789
790
791
792
793
794
795
796
797
(a) (b) 798
Fig. 11 For Scenario 3, when the fluctuations do not exist in traffic demands, sample 799
intersection view in time same time interval in case of implementing (a) the optimum fixed-800
time management (OFTM), (b) vehicle-actuated management (VAM4). 801
802
803
Scenario 3
West West
North North
East East
South South
VAM4 OFTM
Simulation Time 1475 sec.
Simulation Time 1475 sec.
VAM4 OFTM
West West
South South
North North
East East
Simulation Time 1475 sec.
Simulation Time 1475 sec.
Scenario 3
![Page 37: Design of an Efficient Vehicle-Actuated Signal Control ...scientiairanica.sharif.edu/article_22418_3cb77f0f300eb1046401b00b… · 1 1 Design of an Efficient Vehicle-Actuated Signal](https://reader035.vdocuments.mx/reader035/viewer/2022070221/61357ca30ad5d206764768b1/html5/thumbnails/37.jpg)
37
804
805
806
Fig. 12 Obtained reduction rates in case of implementing of vehicle-actuated management 807
instead of optimum fixed-time management for each performance criteria. 808
809
810
811
0%
10%
20%
30%
40%
50%
60%
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Red
uct
ion
Ra
tes
Scenario No
Average Vehicle Delay
Demand: Not-Fluctuated
Demand: Fluctuated
0%
10%
20%
30%
40%
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Red
uct
ion
Ra
tes
Scenario No
Fuel Consumption
Demand: Not-Fluctuated
Demand: Fluctuated
0%
10%
20%
30%
40%
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Red
uct
ion
Ra
tes
Scenario No
Exhaust Emission
Demand: Not-Fluctuated
Demand: Fluctuated
![Page 38: Design of an Efficient Vehicle-Actuated Signal Control ...scientiairanica.sharif.edu/article_22418_3cb77f0f300eb1046401b00b… · 1 1 Design of an Efficient Vehicle-Actuated Signal](https://reader035.vdocuments.mx/reader035/viewer/2022070221/61357ca30ad5d206764768b1/html5/thumbnails/38.jpg)
38
Table 1 Minimum green times for different detector locations 812
Option O-1 O-2 O-3 O-4 O-5 O-6 O-7
Distance between detectors and stop line (m) 15 20 25 30 35 40 45
Minimum green times (sec) 5 7 9 11 13 15 17
*O-1, 2, …..7 shows the options of distance between detector and stop line. 813
814
815
816
Table 2 Reference demand matrix created for analysis studies. 817
Volumes (veh/hr) Flow Rates
Intersection
Approach
(O-D)
West North East South
Intersection
Approach
(O-D)
West North East South
West - 200 700 100 West - 0.200 0.700 0.100
North 150 - 100 150 North 0.375 - 0.250 0.375
East 600 100 - 150 East 0.706 0.118 - 0.176
South 200 250 200 - South 0.308 0.384 0.308
West Intersection Approach: 1000 veh/hr
North Intersection Approach: 400 veh/hr
East Intersection Approach: 850 veh/hr
South Intersection: 650 veh/hr
818
819
820
Table 3 The objective function, the decision variables and the set of constraints for the signal 821
timing optimization problem. 822
Objective
Function ( )f x
n = 1, 2, …,14
( n = Total number of lanes at
intersection)
( )1 1
n n
i i
i i
f x min d D q= =
= =
Decision
Variables
i = 1, 2, …,4
( i = Phase no) ig : Green signal timing (green
split) for phase i (sec)
The Set of
Constraints
Constraints for minimum and maximum
green signal timing ( ig )
i = 1, 2, …,4
( i = Phase no)
7 45ig
Constraints for degree of saturation ( nx )
n = 1, 2, …,14
( n = Total number of lanes at
intersection)
0 1 2nx .
Total Traffic Flow at Intersection: 2900
veh/hr
![Page 39: Design of an Efficient Vehicle-Actuated Signal Control ...scientiairanica.sharif.edu/article_22418_3cb77f0f300eb1046401b00b… · 1 1 Design of an Efficient Vehicle-Actuated Signal](https://reader035.vdocuments.mx/reader035/viewer/2022070221/61357ca30ad5d206764768b1/html5/thumbnails/39.jpg)
39
Table 4 Optimum signal timings and average vehicle delays for created traffic scenarios 823 D
em
an
d
Scen
ario
No
Ph
ase
I (
sec)
Ph
ase
II
(sec
)
Ph
ase
III
(s
ec)
Ph
ase
IV
(se
c)
Cy
cle
Tim
e
(sec
)
Av
era
ge
Veh
icle
Dela
y
(sec
/veh
)
Dem
an
d
Scen
ario
No
Ph
ase
I (
sec)
Ph
ase
II
(sec
)
Ph
ase
III
(s
ec)
Ph
ase
IV
(se
c)
Cy
cle
Tim
e
(sec
)
Av
era
ge
Veh
icle
Dela
y
(sec
/veh
)
Lo
w
1 10 9 9 12 60 25.00
Hig
h
11 19 9 17 14 79 34.00
2 9 7 8 9 53 20.82 12 25 11 22 17 95 43.82
3 19 8 16 13 76 33.25 13 24 11 21 17 93 44.63
4 13 9 11 11 64 26.89 14 25 12 22 18 97 45.55
5 12 7 10 9 58 23.60 15 31 14 27 22 114 55.59
Mo
dera
te 6 14 8 12 11 65 28.32
Very H
igh
16 45 19 39 31 154 78.83
7 19 8 16 13 76 32.84 17 45 20 39 31 155 82.37
8 19 9 17 13 78 33.30 18 31 15 27 22 115 57.56
9 (R) 19 9 16 13 77 33.20 19 45 20 39 31 155 81.21
10 18 9 16 13 76 34.51 20 45 20 39 31 155 80.26
Phase I: West ; Phase II: North / Phase III: East / Phase IV: South
Inter Green Time: 5 sec.
(R): Reference Demand
824
825
Table 5 Scenario-based comparison of average vehicle delay values (sec/veh) 826
Dem
an
d
Scen
ario
No
Ak
çeli
k
(sec
/veh
)
VIS
SIM
(sec
/veh
)
Dem
an
d
Scen
ario
No
Ak
çeli
k
(sec
/veh
)
VIS
SIM
(sec
/veh
)
Dem
an
d
Scen
ario
No
Ak
çeli
k
(sec
/veh
)
VIS
SIM
(sec
/veh
)
Dem
an
d
Scen
ario
No
Ak
çeli
k
(sec
/veh
)
VIS
SIM
(sec
/veh
)
Lo
w
1 25.00 25.51
Mo
dera
te
6 28.32 27.52
Hig
h
11 34.00 34.47
Very H
igh
16 78.83 82.44
2 20.82 21.68 7 32.84 33.38 12 43.82 42.32 17 82.37 84.21
3 33.25 33.91 8 33.30 31.74 13 44.63 42.73 18 57.56 56.16
4 26.89 26.60 9 33.20 32.23 14 45.55 44.82 19 81.21 79.88
5 23.60 23.76 10 34.51 34.94 15 55.59 56.10 20 80.26 79.55
827
828
Table 6 Optimum detector placements for created scenarios 829
Scenario
No
Distance between Detectors
and Stop Line (m) Scenario
No
Distance between Detectors
and Stop Line (m)
West
Approach
North
Approach
East
Approach
South
Approach
West
Approach
North
Approach
East
Approach
South
Approach
1 15 15 15 15 11 20 20 20 20
2 15 15 15 15 12 30 30 30 30
3 15 15 15 15 13 30 30 30 30
4 15 15 15 15 14 35 35 35 35
5 15 15 15 15 15 45 45 45 45
6 15 15 15 15 16 45 45 45 45
7 20 20 20 20 17 45 45 45 45
8 20 20 20 20 18 45 45 45 45
9 25 25 25 25 19 45 45 45 45
10 20 20 20 20 20 45 45 45 45
830
![Page 40: Design of an Efficient Vehicle-Actuated Signal Control ...scientiairanica.sharif.edu/article_22418_3cb77f0f300eb1046401b00b… · 1 1 Design of an Efficient Vehicle-Actuated Signal](https://reader035.vdocuments.mx/reader035/viewer/2022070221/61357ca30ad5d206764768b1/html5/thumbnails/40.jpg)
40
Table 7 For the cases of fluctuated and not-fluctuated traffic demands, obtained 831
reduction rates (success rates) 832
Performance Criteria
For created scenarios, obtained reduction rates (%)
in case of implementing of VAM4 instead of OFTM
Demand: Not - Fluctuated Demand: Fluctuated
Minimum Maximum Average Minimum Maximum Average
Average vehicle delay 4.51 29.16 14.20 15.58 58.55 32.54
Fuel Consumption 2.80 17.68 7.29 9.62 38.41 21.83
Exhaust Emission 2.79 17.68 7.29 9.60 38.40 21.82
833
834
A Brief Technical Biography of Authors 835
Ziya CAKICI received B.Sc. degree from Civil Engineering Department of Celal Bayar 836
University, Manisa, Turkey in 2010. He received the M.Sc. and Ph.D. degrees in transportation 837
engineering from Graduate School of Natural and Applied Sciences at Pamukkale University 838
(Turkey), in 2014 and 2020, respectively. He has been working as Research Assistant in 839
Department of Civil Engineering at Bayburt University. He studies in the fields of traffic 840
management (especially Intelligent Transportation Systems), traffic and transportation 841
planning and traffic optimization. He has published several papers in different journals and 842
international conference proceedings. 843
844
Yetis Sazi MURAT received B.Sc. degree from Civil Engineering Department of Dokuz Eylul 845
University, Izmir, Turkey in 1992. He has received M.Sc. degree from Pamukkale University 846
in 1996 and Ph.D. degree from Istanbul Technical University in 2001. He has been working as 847
Professor in Department of Civil Engineering at Pamukkale University. He worked as visiting 848
scholar at Virginia Tech. University, Falls Church, USA in 2006 (6 months) and University of 849
Nevada, Reno, USA in 2017. He has published many technical-scientific research papers and 850
conducted many projects on different subjects of Transportation Engineering. His main 851
research interests include fuzzy logic and neural network based traffic management approaches. 852
He studies in the fields of traffic management (especially Intelligent Transportation Systems), 853
traffic and transportation planning, traffic accidents and etc. 854
855
Metin Mutlu AYDIN received B.Sc. degree from Civil Engineering Department of Dokuz 856
Eylul University, Izmir, Turkey in 2010. He received the M.Sc. degree from Dokuz Eylul 857
University in 2012 and Ph.D. degree from Akdeniz University, Antalya, Turkey in 2017. He 858
![Page 41: Design of an Efficient Vehicle-Actuated Signal Control ...scientiairanica.sharif.edu/article_22418_3cb77f0f300eb1046401b00b… · 1 1 Design of an Efficient Vehicle-Actuated Signal](https://reader035.vdocuments.mx/reader035/viewer/2022070221/61357ca30ad5d206764768b1/html5/thumbnails/41.jpg)
41
has been working Associate Professor in Department of Civil Engineering at Ondokuz Mayıs 859
University. He studies in the fields of traffic and transportation engineering. His research 860
interests include modelling of the lane discipline, optimizing of the intersection geometry, 861
simulating of driver behaviours, in generally. He has published many research papers in 862
different journals and international conference proceedings. 863
864