Plane Thermoelastic Waves in Infinite Half-Space Caused


Operational Research in Engineering Sciences: Theory and Applications 
Vol. 1, Issue 1, 2018, pp. 40-61 
ISSN: 2620-1607 
eISSN: 2620-1747 

 DOI: https://doi.org/10.31181/oresta19012010140s 

E-mail address: mirko.stojcic@sf.ues.rs.ba, mirkostojcic1@hotmail.com 

APPLICATION OF THE ANFIS MODEL IN ROAD TRAFFIC 
AND TRANSPORTATION: A LITERATURE REVIEW FROM 

1993 TO 2018  

Mirko Stojčić 

Faculty of Transport and Traffic Engineering Doboj, University of East Sarajevo, Vojvode 
Mišića 52 Doboj, Bosnia and Herzegovina  

 

Received: 19 September 2018 

Accepted: 02 November 2018 

Published: 19 December 2018 

 
Review paper 

Abstract: The paper’s focus is on researching the application of the ANFIS (Adaptive 
Neuro Fuzzy Inference System) model in traffic and transport through a review of 
relevant papers. The ANFIS, as an element of artificial intelligence, is widely used in 
intelligent transport systems. All collected papers are divided into 7 sub-areas, namely: 
1) vehicle routing, 2) traffic control at intersections with light signaling, 3) vehicle 
steering and control, 4) safety, 5) modeling of fuel consumption, engine performance 
and exhaust emissions, 6) traffic congestion prediction, and 7) other applications. For 
each sub-area, the analysis of the proposed models is performed with a tabular 
overview of respective input and output variables, while in the third section the 
discussion of the results is given. It is found that the steering and control of vehicles 
represent a sub-area with the highest percentage in the total number of examined 
papers, while the security applications take second place. 

Key Words: ANFIS, Intelligent Transportation Systems, Light Signaling, Vehicle 
Routing, Prediction, Modeling 

1 Introduction 

The development of science and technology has affected a wider study as well 
as application of the solutions based on artificial intelligence in various areas. 
Intelligent transportation systems represent a scientific and engineering discipline 
that implies integration of modern information and communication technologies into 
transport infrastructure and vehicles. Therefore, it is evident that smart solutions 
find their application in this field as well. Some of the most commonly used elements 
of artificial intelligence are fuzzy logic, artificial neural networks, and genetic 
algorithms. In addition, there are popular combinations of techniques such as: neuro-
fuzzy systems, genetic fuzzy systems, and genetic programming neural networks (Kar 
et al., 2014).  



Application of ANFIS model in road traffic and transportation: a literature review from 1993 to 
2018 

 

41 
 

 

Often, though, a realistic system cannot be modeled precisely due to either 
insufficient or unclear information. When that happens, the solutions based on 
traditional computer methods do not yield satisfactory results. Therefore, an 
emphasis is placed on the neuro-fuzzy systems, which represent integration of fuzzy 
logic and artificial neural networks. Fuzzy logic is an extension of the classical logic so 
that the variables can have a certain degree of belonging to either true or false. The 
basic elements for the processing of ambiguities and uncertainty in the fuzzy logic are 
the fuzzy sets which are mathematically represented by the membership functions. 
The fuzzy technologies are human-oriented, which means they simulate the human 
way of thinking and conclusion-making based on the linguistic variables, which are 
represented by fuzzy sets that linguistic expressions are associated with. In addition 
to the advantages, some of which are already mentioned, the disadvantage of the 
fuzzy logic is the impossibility of its adaptation. This problem is solved by artificial 
neural networks representing models of the human brain with interconnected basic 
process elements - artificial neurons. The main features that distinguish them are the 
ability to learn from examples and adaptability, which is characteristic of man, as well 
as in the case of the fuzzy logic (Arora & Saini, 2014). Each neural network is defined 
with three properties: the type of artificial neurons, i.e. the type of their transfer 
function, the connection between the nodes and their structure, and the training 
algorithm. It can be said that the fuzzy logic and the artificial neural networks 
complement each other. One of the most commonly used neuro-fuzzy systems is an 
adaptive neuro-fuzzy inference system (ANFIS), first introduced by Jang 1993 (Jang, 
1993). The problems that have led to the development of the ANFIS are the lack of a 
unique methodology that would transform human knowledge into the base of fuzzy 
rules, as well as the need for a method that will provide, for certain inputs, the 
minimum deviation of outputs from the expected values. The ANFIS model is trained 
by the input-output pairs (vectors), which adjusts the parameters of the membership 
functions of the input and output variables (Jang, 1996). The training algorithm is 
hybrid and combines the gradient descend method and the Last Square Estimation 
(LSE). The fuzzy inference is based on the Takagi-Sugeno system whose typical rule 
has the form: IF A THEN B, where A and B fuzzy sets are described by the 
membership functions. The ANFIS has a five-layer structure, and the network is a 
feed-forward type where neurons transmit their outputs to neuron inputs in the next 
layer and so on, without a cycle. The most important applications of the observed 
neuro-fuzzy model are the modeling of non-linear systems, chaotic time series 
prediction, and clustering. 

The main objective of the survey is to review the ANFIS application in the field 
of road traffic and transportation, as components of the intelligent transportation 
systems. By searching the Web and the Google Scholar bibliographic database, the 
papers that deal with this topic since 1993 have been collected. All papers are divided 
into 7 sub-areas, namely: 1) vehicle routing, 2) traffic control at intersections with 
light signaling, 3) vehicle steering and control, 4) safety, 5) modeling of fuel 
consumption, engine performance and exhaust emissions, 6) traffic congestion, and 
7) other applications. 

Following the introduction, the paper is structured in three sections. The 
literature review deals with the analysis of papers in individual sub-areas with 
tabular representations of the variables of the proposed ANFIS models. The 



Stojčić/Oper. Res. Eng. Sci. Theor. Appl. 1 (1) (2018) 40-61 

 

42 
 

discussion is in the third section where the statistical review of the papers by the year 
of publication is given. On the basis of everything stated in the paper, the last section 
gives a conclusion. 

2 Literature review 

2.1 Vehicle routing 

An increasing number of vehicles on roads, especially in cities, are causing 
great traffic jams as existing roads do not have the required capacity. In order to 
avoid or mitigate this problem, the choice of the optimal route of the vehicle is a very 
big challenge. Apart from avoiding traffic jams, the optimal route is selected on the 
basis of several criteria, some of which may be: travel time, distance, fuel 
consumption, road works, etc. It is evident that it is very difficult to find a route that 
meets all the requirements. Abbas et al. (2011) propose a model that represents 
integration of artificial neural networks, a neuro-fuzzy model, and an ant colony 
optimization algorithm to select the optimal route. All necessary input data are 
provided by the traffic control center. The proposed model is capable of dynamically 
adjusting the route change. 

The choice of route for transport of dangerous goods in the city is a very 
complex task. In (Pamucar et al., 2016), a modified ANFIS model with the Dijskstra 
algorithm for determining the optimal route is proposed, i.e. ANFIS-D model. After 
training with the artificial bee colony algorithm, for the new input data, the model 
gives the value of the cost-risk ratio for each branch of the network individually. The 
role of the Dijkstra algorithm is to find a route in the network that minimizes the total 
value of the given ratio. The described model was tested in (Pamucar et al., 2016a) in 
the selection of optimal routes for the transport of oil and oil derivatives in Belgrade, 
Serbia. 

Similarly to the described model, Pamučar & Ćirović (2018) represent the 
ANFIGS (Adaptive Neuro Fuzzy Inference Guidance System) model for choosing the 
route of vehicle movement under the conditions of uncertainty. In the neuro-fuzzy 
system the knowledge of the dispatcher is accumulated and seven criteria are defined 
that influence the selection of the route. The clustering technique is applied in the 
paper. One of the main advantages of this model is its ability to dynamically adapt to 
unpredictable events on the route.  

In the conditions of natural disasters, it is very important to respond quickly 
and provide assistance to the affected areas as soon as possible. Under such 
conditions, the roads are often damaged but other factors that adversely affect the 
rapid route planning appear as well. Gharib et al. (2018) use the ANFIS in the first 
step of selecting a route for classification of critical areas into two clusters: 1) areas 
that can be assisted by road and 2) areas with an access only from the air. Table 1 
shows the input and output variables for the listed ANFIS models. 

 

 

 



Application of ANFIS model in road traffic and transportation: a literature review from 1993 to 
2018 

 

43 
 

 

Table 1 Input and output variables of the ANFIS models for vehicle routing 

Author/year Input variables Output variables 

Abbas et al. 
(2011) 

Distance; traffic flow; environment monitoring; 
width; road condition; traffic lights 

Pheromone level 
(ant colony) 

Pamučar et al. 
(2016), 

Pamučar et al. 
(2016a) 

Carrier’s operating costs; emergency 
response; risk associated with the 

environment; risk of an accident; consequences 
of an accident; risk associated with 

infrastructure; risks of terror attack/hijack 

Cost/risk value 

Pamučar & 
Ćirović (2018) 

Type of road surface; travel distance; travel 
time; route capacity; traffic capacity; 

road capacity; the existence of alternative roads 
along the length of the route 

Preference of the 
dispatcher to 

select a particular 
route 

Gharib et al. 
(2018) 

Road slope; weather conditions; intensity of 
disaster; population density; road 
risk; distance of vehicle; distance 

from airport; road width 

Cluster (1 or 2) 

2.2 Traffic control at intersections with light signaling 

The application of light signalization to control traffic at intersections is one of 
the most common and most effective methods. However, a great lack of this kind of 
regulation is that the intervals are fixed, which can often cause unnecessary delays 
and congestions. An intelligent solution consists in forming an adaptive model that 
adjusts intervals to the real state of traffic at the intersection. Such a model is 
presented in (Udofia et al., 2014). Its basis is the ANFIS model with two inputs. For 
training data, the urgency degree as an output variable is calculated analytically 
based on the input variables for each phase of the crossroads individually. The model 
uses real data collected by the sensor and gives a certain output based on them. The 
next green interval is assigned to the phase with the highest urgency degree. The 
model was tested at a real intersection and the results confirm its effectiveness. The 
described ANFIS model is also used in (George et al., 2015) within a system that 
receives incoming traffic data from the processing of video data. Lai et al. (2015) also 
use the same inputs, while the output variable is an extension time of the duration of 
the green light interval. The testing has found that the performance of the proposed 
model is better than that of the traditional and fuzzy controllers. The ANFIS traffic 
control model can also be tested using the graphical user interface in the MATLAB 
software package, which was done in (Abiodun et al., 2014). The model proposed in 
(Wannige & Sonnadara, 2008) has two inputs representing the number of vehicles 
entering the intersection in both directions. The model training was performed for 
the given input values and for calculating an optimal time of the green light interval 
based on them. According to Seesara & Gadit (2015), two input variables were 
selected based on the advice of competent institutions and traffic experts. In this 
paper, comparison of performance is performed between the ANFIS and the fuzzy 
controller with the ANFIS giving better results. Arraghi et al. (2014) observe four 



Stojčić/Oper. Res. Eng. Sci. Theor. Appl. 1 (1) (2018) 40-61 

 

44 
 

input variables for traffic control at the four-way intersection. In this case, the ANFIS 
justifies the application because it shows better testing results than fuzzy controllers 
and fixed-time models. Korkmaz & Akgüngör (2016) use the ANFIS to model vehicle 
delays at vehicle intersection, and, according to Gokdag et al. (2007), a model of the 
same purpose has a different set of input variables. Comparison of the prediction 
results with those of the usual methods, such as Webster, HCM, DDF and SSM, 
indicates that the ANFIS represents a very promising modeling method. Testing was 
carried out in the case of an intersection in Erzurum, Turkey. Similar research is also 
presented in (Hasiloglu et al., 2014), where comparison was done, instead of the DDF, 
with the Multiple Regression Analyzes method. The observed variables are the same 
for the two above mentioned models. Table 2 provides an overview of the input and 
output variables of the ANFIS models by authors. 

Table 2 Input and output variables of the ANFIS model for traffic control at 

intersections with light signaling 

Author/year Input variables Output variables 

Udofia et al. 
(2014) 

Waiting time; queue length Urgency degree 

George et al. 
(2015) 

Waiting time; queue length Urgency degree 

Lai et al. 
(2015) 

Waiting time; queue length 
Extension 

time of the next phase 

Abiodun et al. 
(2014) 

Number of vehicles on the arrival side; 
number of vehicles on the queuing side 

Extension time of green 
light 

Wannige & 
Sonnadara 

(2008) 
Vehicle inflow in two roads 

Green light time of 
one lane 

Seesara & 
Gadit (2015) 

Arrival rate of the particular phase; last 
time vehicles that have not passed during 

last green phase 
Green time extension 

Arraghi et al. 
(2014) 

Queue length of vehicles at each 
approaching link (for 4 links) 

Green time for the 
current phase 

Gokdag et al. 
(2007) 

Time; number of approaching vehicles in 
the green duration; number of queuing 

vehicles in the red duration 
Vehicle delay 

Korkmaz & 
Akgüngör 

(2016) 

Cycle time of signalization; 
green time; degree of saturation 

Vehicle delay 

Hasiloglu et al. 
(2014) 

Time; number of approaching vehicles in 
the green duration; number of queuing 

vehicles in the red duration 
Vehicle delay 



Application of ANFIS model in road traffic and transportation: a literature review from 1993 to 
2018 

 

45 
 

 

2.3 Vehicle steering and control 

A large number of controllers for control and stability in vehicles are based on 
neuro-fuzzy systems. Selma & Chouraqui (2012) propose ANFIS models to control 
vehicle paths based on previous training. Two models for positioning the X and Y axis 
have been developed. The model was tested by the simulation method and the results 
show its efficiency. According to Saifizul et al. (2006), the ANFIS model for steering 
has the task of keeping the lateral error and the yaw error at an acceptable level 
while driving. In this case, the input data are collected by means of camera on-board, 
which is a much simpler solution than the existing ones, which implies the 
installation of a magnet or wiring on the road. The ESC (Electronic Stability Control) 
is an unavoidable system in newer cars that significantly improves passenger safety. 
The ECS systems mainly use measured yaw velocity of the chassis and the sideslip 
angle (the angle between the directions of the vehicle's velocity and its chassis). The 
problem is the determination of the given angle because it is difficult to measure with 
the sensor. A Sideslip angle modeling involves the use of various methods, and Boada 
et al. (2015), propose ANFIS for this purpose. In (Boada et al., 2016), the same author 
uses the Kalman filter to evaluate the Sideslip angle, in combination with the ANFIS 
model. However, Hou et al. (2008) uses the Sideslip angle as one of the input 
variables in the integrated chassis control model. Model training and testing are 
carried out using the simulation method.  

Automatic transmission control in modern vehicles is done with the computer 
that selects the optimal shift based on the input signals received by the sensors. 
However, in some driving conditions such a system is not efficient (low speed, vehicle 
load, etc.). A potential solution is presented in (Li et al., 2007) and is based on the 
ANFIS model. Perez et al. (2010) present an ANFIS model for controlling the braking 
and acceleration of autonomous vehicles that tend to expand in the future. The tests 
confirm the efficiency of the ANFIS model in determining the value of the output 
variables. Autonomous vehicles and the ANFIS model are also studied in (Al Mayyahi 
et al., 2014), where four such models are developed to avoid obstacles and reach the 
desired position.  

When it comes to electric vehicles, using the observed neuro-fuzzy model in 
the regenerative braking system, it is possible to provide greater autonomy (Sindhuja 
et al., 2014). The system involves the use of an electric motor as a generator in 
braking, thus recycling the spent energy into a rechargeable battery. The ANFIS 
model is also applicable in the case of hybrid drives where it minimizes engine fuel 
consumption with internal combustion and maximizes torque (Mohebbi et al., 2005).  
Eski & Yıldırım (2017) describe the use of ANFIS model for the electronic regulation 
of throttle of heavy vehicles. Car parking is a demanding action, sometimes for 
experienced drivers, and if it is a truck with a trailer, the problem becomes very 
complex. Due to the non-linearity of the movement of such a vehicle, the observed 
neuro-fuzzy system was applied by Azadi et al. (2013). In the first stage of the 
proposed model, the vehicle in advance takes an adequate position in order to then 
position it back to the parking place. The use of sensors that provide environmental 
information is unavoidable in this case.  

Several authors dealt with the use of an ANFIS suspension model to improve 
safety and travel comfort (Shuliakov et al., 2015; Nugroho et al., 2014; 



Stojčić/Oper. Res. Eng. Sci. Theor. Appl. 1 (1) (2018) 40-61 

 

46 
 

Kothandaraman & Ponnusamy, 2012). Depending on the input data, the model is 
capable of adapting the characteristics of the shock absorbers and other elements 
that make up the mentioned system. An overview of the input and output variables of 
some ANFIS models with application in vehicle steering and control systems is shown 
in Table 3. 

Table 3 Input and output variables of the ANFIS model in vehicle steering and 

vehicle control systems 

Author/year Input variables Output variables 

Selma & Chouraqui 
(2012) 

X position, Y position X position, Y position 

Saifizul et al. 
(2006) 

Lateral error; angle between 
longitudinal direction and local road tangent 

at look-ahead distance; yaw rate 
Steering angle 

Boada et al. 
(2015); Boada et 

al. (2016) 

Lateral acceleration; yaw rate; steering 
angle; longitudinal velocity; yaw 

rate/longitudinal velocity 
Sideslip angle 

Hou et al. (2008) 
Yaw velocity discrepancy; sideslip angle 

discrepancy 
Brake/Throttle 

Li et al. (2007) Vehicle velocity; air damper angle Shift point 

Perez et al. (2010) Speed error; acceleration Brake/Throttle 

Sindhuja et al. 
(2014) 

Distribution of braking force; 
(front)battery’s state of charge (SOC); speed 

of the motor 
Braking force ratio 

Al Mayyahi et al. 
(2014) 

Angle difference (for the first and second 
controller); front, right and left distance (for 

for the third and fourth controller) 
Right/left angular velocity 

Mohebbi et al. 
(2005) 

Desired torque; battery’s state of charge 
(SOC) 

Throttle angle of the 
internal combustion 

engine 

Eski & Yıldırım, 
(2017) 

Two different random inputs of the heavy 
duty vehicle speed 

Servo motor speed 

Azadi et al. (2013) 
Tractor yaw angle; trailer yaw angle; 

horizontal distance from the wall 
Steering angle 

Shuliakov et al. 
(2015) 

Turn rate; angular transducer output 
Deviation angle of a 
stabilization object 

Nugroho et al. 
(2014) 

Velocity of sprung mass (car body); relative 
velocity between sprung mass and unsprung 

mass/velocity of unsprung mass (wheel); 
relative velocity between the sprung mass 

and unsprung mass 

Fuzzy-skyhook 
force/fuzzy-ground force 

Kothandaraman & 
Ponnusamy, 

(2012) 

Suspension deflection; sprung mass 
velocity 

Actuator 
force 



Application of ANFIS model in road traffic and transportation: a literature review from 1993 to 
2018 

 

47 
 

 

2.4 Safety 

Security has always been the highest priority in traffic, and today a large 
number of technologies (video surveillance, speed control, etc.) are present within 
intelligent transport systems, which have the task of raising safety to an even higher 
level (Rahimi, 2017). Every day, an increasing number of vehicles are in the streets 
and so are drivers who do not share the same experience and abilities. Statistics say 
that the driver's behavior is the main cause of traffic accidents. Bearing this in mind, a 
number of authors have paid attention to the development of various driver behavior 
prediction models, some of which are listed in (Kumar & Prasad, 2015).  

The ANFIS application for the car following model is presented in (Poor et al., 
2016; Khodayari et al., 2010). Similarly, Ghaffari et al. (2015) represent a new 
approach to modeling the car following when changing the lane of the leading vehicle. 
Such a maneuver can be viewed as a transient condition because the vehicle deviates 
from the conventional modeling for a certain time. The same author deals with the 
modeling of the overtaking path in (Ghaffari et al., 2011, 2011a), as one of the most 
demanding traffic operations.  

Modern Collision Avoidance Systems involve the use of various sensors in 
order to collect the data necessary for determining the parameters. All this raises the 
price and complexity of such systems. Bearing this in mind, Saadeddin et al. (2013) 
develop a low-cost system based on a combination of the INS (Inertial Navigation 
System) data and a GPS (Global Positioning System) in their research. This integration 
is realized through the IDANFIS (Input-Delayed ANFIS). The data provided by 
satellite systems have been used as inputs in (Sun et al., 2017) in combination with a 
neuro-fuzzy model to develop a rear-end impact prevention system.  

Dadula & Dadios (2016) represent an ANFIS which has the function of 
detecting critical events in public passenger transport based on characteristic sounds. 
The system can differentiate the normal circumstances from alarming (e.g. shooting) 
with a high percentage of accuracy.  

Pedestrians are a very vulnerable group of participants in the traffic. For the 
sake of their protection, various mechanisms can be implemented in intelligent 
transport systems. One of them is modeling the pedestrian decision to cross the street 
with the help of artificial neural networks and the fuzzy logic, as presented in 
(Ottomanelli et al., 2010).  

Determining critical points along the road can be of great use in preventing 
traffic accidents. In the case that statistical methods cannot provide reliable results, 
e.g. because of the lack of data, the authors use the observed neuro-fuzzy system that, 
based on the physical characteristics of the path and environmental factors, predicts 
the risk spots. Such studies are presented in (Hosseinlou & Sohrabi, 2009; Effati et al., 
2014). Prediction of traffic accidents in real time using ANFIS is presented in (Liu & 
Chen, 2017). The authors analyze the traffic flow factors just before an accident 
occurs. By comparing the results with other models, it can be concluded that the 
ANFIS in this case also shows better performance.mTraffic sign detection is an 
important part of the Driver Assistance System because it allows automatic 
adjustment to the conditions prescribed for them. Billah et al. (2015) propose an 
ANFIS model for the recognition of circular signs based on the data obtained by image 



Stojčić/Oper. Res. Eng. Sci. Theor. Appl. 1 (1) (2018) 40-61 

 

48 
 

processing and video processing. The recognition accuracy is more than 98%, which 
sufficiently highlights the model’s capabilities. 

In order to improve the vehicle stability as well as its handling, it is important 
to adjust the speed to the road geometry. The model that performs this function is 
presented in (Wankhede et al., 2011). Its output represents a certain degree of 
acceleration or deceleration of the vehicle, depending on the current acceleration and 
winding of the road. Table 4 provides an overview of the security applications of the 
ANFIS model with input and output variables. 

Table 4 Input and output variables of the ANFIS model in security 

applications 

Author/year Input variables Output variables 

Poor et al. 
(2016) 

Distance difference (between cars); 
velocity difference; speed of the front car; 

driver reaction time 

Acceleration of following 
vehicle 

Khodayari et al. 
(2010) 

Relative speed; relative distance; acceleration 
of leading vehicle 

Acceleration of following 
vehicle 

Ghaffari et al. 
(2015) 

Distance between follower and front vehicle; 
relative acceleration of these two vehicles; 

velocity of follower; acceleration of follower 

Acceleration of following 
vehicle 

Ghaffari et al. 
(2011) 

Lateral coordinate; longitudinal coordinate; 
velocity; acceleration; movement angle; 

Lateral coordinate; 
longitudinal coordinate 

Ghaffari et al. 
(2011a) 

Velocity; acceleration; jerk; heading angle; 
heading angle race 

Acceleration; 
heading angle 

Saadeddin et al. 
(2013) 

Position and velocity components (x, y, and z 
axis) from INS 

Error in INS position and 
velocity 

Sun et al. 
(2017) 

Relative Distance; relative velocity; 
relative heading 

Warning status 

Dadula & 
Dadios (2016) 

12 mel Frequency Cepstral Coefficients 
(MFCCs) for each audio frame 

Crisis condition or normal 
condition 

Ottomanelli et 
al. (2010) 

Vehicle’s speed; vehicle’s distance; interval 
between vehicle arrival and pedestrian arrival 

at the crossing (or gap) 
Decision (wait or cross) 

Hosseinlou & Topographical and geometrical Accident frequency of the 



Application of ANFIS model in road traffic and transportation: a literature review from 1993 to 
2018 

 

49 
 

 

Sohrabi (2009) drawings of the road; amount of 
traffic volume per day; amount of hourly traffic 

volume in the day 

road 

Effati et al. 
(2014) 

Roadway geometry; environmental 
factors 

Danger value 

Liu & Chen 
(2017) 

Average speed; volume; occupancy in 
30-second aggregation intervals (9 traffic flow 

variables) 
Crash risk value 

Billah et al. 
(2015) 

Total black pixel; entropy; contrast; 
correlation; energy; homogeneity 

Label which means a 
specific sign 

Wankhede et al. 
(2011) 

Angle curvature; acceleration Acceleration 

2.5 Modeling of fuel consumption, engine performance and exhaust emissions 

Fuel consumption in the world is growing rapidly every day, while, at the same 
time, the world reserves are decreasing. In addition to the problem of energy 
shortages, the problem of increasing pollution is present, that is, the problem of 
harmful substances emissions into the atmosphere. Traffic and transport activities 
constitute a very large share of the total fuel consumption, and therefore, studies 
have focused on optimization. To do this, it is necessary to develop models for the 
consumption prediction. The model presented in (Massoud et al., 2014) takes into 
account the interaction of transport and land use in urban areas so that the planners 
can efficiently analyze and plan fuel consumption. When it comes to passenger cars, 
the ANFIS prediction model is proposed in (Syahputra, 2016; Atmaca et al., 2001). 
According to Abdallat et al. (2011), using the given model, it is possible to estimate 
the need for the amount of fuel for the transportation of the whole country. In the 
concrete case, the research was carried out for Jordan. 

Diesel fuel is mostly used for trucks, and in order to reduce CO2 emissions, the 
use of alternative fuels, such as biodiesel, is increasingly considered. Many studies 
deal with analyzing the effects of the addition of diesel fuel. The authors propose 
ANFIS models that have the task of predicting engine performance and concentration 
of harmful substances of exhaust gases when using such mixtures (Hosoz et al., 2013; 
Özkan et al., 2015; Ghanbari et al., 2015; Rai et al., 2015). Table 5 provides an 
overview of the ANFIS model with application in modeling fuel consumption, engine 
performance and exhaust emissions.  

 

 



Stojčić/Oper. Res. Eng. Sci. Theor. Appl. 1 (1) (2018) 40-61 

 

50 
 

Table 5 Input and output ANFIS variables for modeling fuel consumption, 

engine performance and exhaust emissions 

Author/year Input variables Output variables 

Massoud et al. 
(2014) 

Land use; transportation Energy consumption 

Syahputra 
(2016) 

Car weight; year Miles per gallon 

Atmaca et al. 
(2001) 

Car weight; year Miles per gallon 

Hosoz et al. 
(2013) 

Biodiesel content in the fuel; engine speed; 
engine load 

Brake power; brake 
specific fuel 

consumption; brake 
thermal efficiency; 

emissions of HC, CO, 
NO; exhaust gas 

temperature 

Özkan et al. 
(2015) 

Types of engine fuels; injection pressure; 
speed 

Torque; specific fuel 
consumption; air 

consumption; efficiency; 
lambda values 

Ghanbari et al. 
(2015) 

Diesel–biodiesel and nano particles 
blends; speed 

Engine power; torque; 
brake specific fuel 

consumption; emission 
components 

Rai et al. 
(2015) 

Percentage load; percentage liquefied 
petroleum gas; injection timing 

Brake specific energy 
consumption; brake 
thermal efficiency; 

exhaust gas 
temperature; smoke 

Abdallat et al. 
(2011) 

Annual number of vehicles; vehicle owner 
level; income level; fuel prices 

Energy consumption (in 
tons of oil) 

2.6 Traffic congestion prediction 

Traffic congestion is a part of everyday life in big cities, which has a negative 
impact on life quality because of considerable time spent. In addition to time 



Application of ANFIS model in road traffic and transportation: a literature review from 1993 to 
2018 

 

51 
 

 

expenditure, it is necessary to consider higher fuel consumption, which means more 
air pollution. Due to a number of problems caused by traffic jams, intelligent 
transportation systems should provide mechanisms to anticipate and avoid them 
(Joshi & Hadi, 2015). Zaki et al. (2016) present a framework for short-term 
prediction, where, apart from the ANFIS, a model based on the Hidden Markov 
Models is being developed. The same variables are taken into account by Shancar et 
al. (2012) in their model. Kukadapwar & Parbat (2015) represent an ANFIS model 
that uses real-time traffic data for the prediction of jams in Nagpur city, India. An 
overview of these models is given in Table 6. 

Table 6 Input and output variables of the ANFIS model for predicting traffic 

congestion 

Author/year Input variables Output variables 

Zaki et al. 
(2016) 

Speed; density Level of congestion 

Kukadapwar & 
Parbat (2015) 

Speed reduction rate; proportion of time 
traveling at very low speed (below 5 

kmph) compared with total travel time; 
traffic volume to roadway capacity ratio 

Congestion index 

Shankar et al.  
(2012) 

Speed; density Level of congestion 

2.7 Other applications 

For the purpose of surveillance, future planning and efficient management of 
the transportation system of a country, it is necessary to have accurate data on the 
classes and number of vehicles. Intelligent transportation systems include various 
technologies, and the observed neuro-fuzzy vehicle classification system is proposed 
in (Maurya & Patel, 2015). The authors take into account the physical dimensions of 
the vehicle, such as the wheelbase and the average distance of the wheels on the same 
track. 

Vehicle activated signs to warn drivers of over speeds are a very useful 
mechanism for intelligent transportation systems. However, if the threshold of speed 
is adapted to the conditions and dynamics of traffic, the benefits become even greater 
(Jomaa et al., 2015).  

The prediction of travel time can be realized mostly on the basis of statistics or 
artificial neural networks. Statistical solutions often do not yield satisfactory results 
due to the non-linear nature of the dependencies of the observed variables. 
Therefore, the application of neural networks, more precisely the ANFIS model is 
more appropriate in this case (Maghsoudi & Moshiri, 2017). 

Thipparat & Thaseepetch (2012) propose an ANFIS model for predicting the 
possibility of sustainability of the highway construction. At the design and planning 
stage, expert knowledge is collected in order to evaluate some of the influential 
factors and, based on this, deduce the conclusion on sustainability.  



Stojčić/Oper. Res. Eng. Sci. Theor. Appl. 1 (1) (2018) 40-61 

 

52 
 

The selection of an optimal vehicle for transportation in the Serbian army 
based on a given neuro-fuzzy model was presented in (Pamučar et al., 2013). The 
model is capable of simulating the decision-making process, as do logistics officers. 

In (Ghaffari et al., 2012), the subject of research is a prediction of the future 
status of the vehicle with the Stop&Go system. The developed model can reduce the 
likelihood of impact on the rear of the vehicle, and in addition, improve the comfort 
experience during city driving. 

Since traffic is an important source of noise, Sharma et al. (2014) present an 
ANFIS model for predicting the value of the mentioned variable. Vehicle speeds, 
traffic flows and the use of siren can be listed as the main influencing factors. Table 7 
provides an overview of the ANFIS model with applications in intelligent 
transportation systems. 

Table 7. Input and output variables of the ANFIS model for various 

applications in intelligent transportation systems 

Author/year Input variables Output variables 

Maurya & 
Patel (2015) 

Wheelbase; average track 

Light commercial 
vehicle/ car-jeep-

van/two axle trucks-
bus/three axle truck/ 

multi axle trucks 

Jomaa et al. 
(2015) 

Time of day; traffic flow; standard 
deviation of mean vehicle speeds 

85th percentile speed for 
each hour on the day 

Maghsoudi & 
Moshiri 
(2017) 

Vehicle speed; road occupation coefficient; 
traffic flow 

Travel time 

Thipparat & 
Thaseepetch 

(2012) 

Geometrics and alignments; earthworks; 
pavement; drainage; retaining walls; slope 

protection; landscape and ecology… (14 
groups, 60 variables) 

Sustainability level of 
highway design 

Pamučar et al. 
(2013) 

Reliability of the means of transport; 
mobility of the means of transport in field 
conditions; exploitation of the cubage of 

transport; cost of tonal kilometer 

Preferential dispatcher 

Ghaffari et al. 
(2012) 

Relative speed; relative distance; 
acceleration of follower vehicle; velocity of 

follower vehicle 

Acceleration of follower 
vehicle in next steps 

Sharma et al. 
(2014) 

Road traffic flow; vehicle speed; honking Traffic noise 



Application of ANFIS model in road traffic and transportation: a literature review from 1993 to 
2018 

 

53 
 

 

3 Discussion 

The adaptive neuro-fuzzy inference system provides wide application in road 
traffic and transportation. In this review, 62 papers were collected for a period of 25 
years of its study. Fig. 1 shows the number of papers published per year. It can be 
concluded that the application of the ANFIS in the observed area was not the subject 
of research until 2001, followed by a break until 2005. Since then, the number of 
papers per year has grown exponentially in order to record the highest value in 2015. 
Nevertheless, in the last few years, there has been a clear decrease in interest in 
studying the given topic. 

 

Fig. 1 Number of papers per years 

The collected papers are divided into 7 sub-areas, as already discussed in 
Section 2. Fig. 2 shows percentage share of the papers from each sub-area in the total 
number. It is obvious that the vehicle steering and control make up the largest 
percentage, 24%, and the safety is immediately behind with 23%.  

Ultimately, the ANFIS application in the area of vehicle steering and control, in 
addition to driving comfort, aims at increasing passenger safety. The smallest number 
of authors dealt with predicting traffic congestion with the help of the observed 
neuro-fuzzy model. 

 



Stojčić/Oper. Res. Eng. Sci. Theor. Appl. 1 (1) (2018) 40-61 

 

54 
 

 

Fig. 2 Participation of individual sub-areas in the total number of collected 

papers 

Table 8 gives an overview of the number of papers by individual areas and by 
year of publication, with the years with no posts left in Table. If observed in 2015, the 
greatest number of papers is published from the sub-area of traffic control at 
intersections with light signaling and modeling of fuel consumption, engine 
performance and exhaust emissions. The second year in terms of the number of 
published papers is 2014 where the largest number of papers is from the sub-area of 
traffic control at intersections with light signaling. 

Table 8 Number of papers by sub-areas and year of publication 

Year VR TC SC S MF CP OA 
2001     1   
2005   1     
2006   1     
2007  1 1     
2008  1 1     
2009    1    
2010   1 2    
2011 1   3 1   
2012   2   1 2 
2013   1 1 1  1 
2014  4 3 1 1  1 
2015  3 2 2 3 1 2 
2016 2 1 1 2 1 1  
2017   1 2   1 
2018 2       

* VR – vehicle routing; TC – traffic control at intersections with light signaling; SC –
vehicle steering and control; S – safety; MF – modeling of fuel consumption, engine 
performance and exhaust emissions; CP – traffic congestion prediction; OA – other 
applications 



Application of ANFIS model in road traffic and transportation: a literature review from 1993 to 
2018 

 

55 
 

 

The total number of the sources dealing with the topic 60, comprising 41 journals 
and 19 conferences. When it comes to the number of papers published by a single 
source, only two magazines have two published papers, namely, 

 Mechanical Systems and Signal Processing, and 

 International Journal of Scientific and Engineering Research.  

Depending on the purpose of the ANFIS model itself, authors use different 
input and output variables, but in a single sub-area, there are many cases in which 
they have opted for the same. The sets of values of the observed variables are 
obtained mainly in two ways, which are measurements and simulation methods in 
one of the softwares. Also, model testing and validation are in many cases performed 
in a simulation environment. For the functioning of ANFIS in real systems, such as 
road vehicles and generally intelligent transportation systems, sensors play a key role 
in providing input data. Model outputs are forwarded as information to the user or 
used as an input of an actuator or a separate system that needs to perform a 
particular action. 

The basic limitation of this paper is the possibility of not including or failing to 
find all the referential papers from the observed area. In addition, papers in non-
English languages are not taken into consideration. 

Conclusions 

The paper analyzes the application of the ANFIS model in the field of road 
traffic and transportation. It presents an overview of the papers, while the proposed 
models for specific purposes are theoretically analyzed with the results tabulated and 
graphically presented. It can be concluded that the use of ANFIS in traffic is largely 
due to its ability to model non-linear systems ans well as its ability of adaptability 
(learning from examples). A key step in developing the ANFIS model is the correct 
choice of input variables depending on the desired output. In addition, in order for 
the model to be trained, it is necessary to collect adequate data. The results of the 
testing of the observed model show its superiority in comparison to the classical, 
previously used models. Some authors combine the ANFIS with other techniques; 
hence, such modified models as ANFIS-D and ANFIGS. Given that the field of 
intelligent transport systems develops every day, new opportunities for potential 
applications of ANFIS are being created. Sensors for data acquisition have a very 
important role as the goal is to provide accurate inputs to the model. Future research 
could aim at analyzing the ANFIS model application to other modes of transport. 

References  

Abbas, S., Khan, M. S., Ahmed, K., Abdullah, M., & Farooq, U. (2011). Bio-inspired 
neuro-fuzzy based dynamic route selection to avoid traffic congestion. International 
Journal of Scientific and Engineering Research, 2(6), 284-289. 

Abiodun, B. A., Amosa, A. A., Olayode, A. O., Morufat, A. T., & Adedayo, S. A. (2014). 
Traffic Light Control Using Adaptive Network Based Fuzzy Inference System. In 



Stojčić/Oper. Res. Eng. Sci. Theor. Appl. 1 (1) (2018) 40-61 

 

56 
 

Proceedings of 2014 International Conference on Artificial Intelligence & 
Manufacturing Engineering (ICAIME 2014), Dubai, December 25-26, 2014 (pp. 156-
161). 

Abdallat, Y., Al-Ghandoor, A., Samhouri, M., Al-Rawashdeh, M., & Qamar, A. (2011). 
Jordan Transport Energy Demand Modeling: The Application of Adaptive Neuro-
Fuzzy Technique. International Review of Mechanical Engineering, 5(7), 1321-1326. 

Al-Mayyahi, A., Wang, W., & Birch, P. (2014). Adaptive neuro-fuzzy technique for 
autonomous ground vehicle navigation. Robotics, 3(4), 349-370. 
https://doi.org/10.3390/robotics3040349 

Araghi, S., Khosravi, A., & Creighton, D. (2014). ANFIS traffic signal controller for an 
isolated intersection. In Proceedings of the International Conference on Fuzzy 
Computation Theory and Applications, Rome, Italy, October 22-24 (pp. 175-180). 
INSTICC Press 

Arora, N., & Saini, J. R. (2014). A literature review on recent advances in neuro-fuzzy 
applications. International Journal of Advanced Networking and Applications, 1, 14-
20. 

Atmaca, H., Cetisli, B., & Yavuz, H. S. (2001). The comparison of fuzzy inference 
systems and neural network approaches with ANFIS method for fuel consumption 
data. In Second International Conference on Electrical and Electronics Engineering 
(ELECO), Bursa, Turkey, November 7-11. 

Azadi, S. H., Nedamani, H. R., & Kazemi, R. (2013). Automatic Parking of an Articulated 
Vehicle Using ANFIS. Global Journal of Science, Engineering and Technology, 14, 93-
104. 

Billah, M., Waheed, S., Ahmed, K., & Hanifa, A. (2015). Real Time Traffic Sign Detection 
and Recognition using Adaptive Neuro Fuzzy Inference System. Communications on 
Applied Electronics, 3(1), 1-5. 10.22068/ijae.7.1.2350 

Boada, B. L., Boada, M. J. L., Gauchía, A., Olmeda, E., & Díaz, V. (2015). Sideslip angle 
estimator based on ANFIS for vehicle handling and stability. Journal of Mechanical 
Science and Technology, 29(4), 1473-1481. https://doi.org/10.1007/s12206-015-
0320-x 

Boada, B. L., Boada, M. J. L., & Diaz, V. (2016). Vehicle sideslip angle measurement 
based on sensor data fusion using an integrated ANFIS and an Unscented Kalman 
Filter algorithm. Mechanical Systems and Signal Processing, 72, 832-845. 
https://doi.org/10.1016/j.ymssp.2015.11.003 

Dadula, C., & Dadios, E. (2016). Event Detection Using Adaptive Neuro Fuzzy 
Inference System for a Public Transport Vehicle. In 11th International Conference of 
the Eastern Asia Society for Transportation Studies. 

Effati, M., Rajabi, M. A., Samadzadegan, F., & Shabani, S. (2014). A geospatial neuro-
fuzzy approach for identification of hazardous zones in regional transportation 
corridors. International Journal of Civil Engineering, 12(3), 289-303. 
http://ijce.iust.ac.ir/article-1-825-en.html 

Eski, İ., & Yıldırım, Ş. (2017). Neural network-based fuzzy inference system for speed 
control of heavy duty vehicles with electronic throttle control system. Neural 
Computing and Applications, 28(1), 907-916. https://doi.org/10.1007/s00521-016-
2362- 



Application of ANFIS model in road traffic and transportation: a literature review from 1993 to 
2018 

 

57 
 

 

George, A. M., Kurian, C. P., George, V. I., & George, M. A. (2015). Hardware in Loop 
Simulation of an Adaptive Traffic Light Control System. International Journal of 
Applied Engineering Research, 10(19), 39999-40004. 

Gharib, Z., Bozorgi-Amirib, A., Tavakkoli-Moghaddamb, R., & Najaa, E. (2018). A 
cluster-based emergency vehicle routing problem in disaster with reliability. Scientia 
Iranica, 25(4), 2312-2330. 10.24200/SCI.2017.4450 

Ghaffari, A., Khodayari, A. R., Kamali, A., & Tajdari, F. A (2015). A New Model of Car 
Following Behavior Based on Lane Change Effects using Anticipation and Evaluation 
Idea. Iranian Journal of Mechanical Engineering, 16(2), 26-38. 

Ghaffari, A., Alimardani, F., Khodayari, A., & Sadati, H. (2011a). ANFIS based modeling 
for overtaking maneuver trajectory in motorcycles and autos. In Proceedings of IEEE 
International Conference on Control System, Computing and Engineering, Penang, 
Malaysia, November 25-27 (pp. 68-73). IEEE 10.1109/ICCSCE.2011.6190498 

Ghaffari, A., Khodayari, A., & Arvin, S. (2011). ANFIS based modeling and prediction 
lane change behavior in real traffic flow. In International Conference on Intelligent 
Computing and Intelligent Systems, Penang, Malaysia, November 25-27 (pp. 156-
161). 

Ghaffari, A., Alimardanii, F., & Khodayari, A. (2012). Predicting the future state of a 
vehicle in a stop&go behavior based on ANFIS models design. In Proceedings of 6th 
IEEE International Conference Intelligent Systems, Sofia, Bulgaria, September 6-8 (pp. 
368-373). IEEE 10.1109/IS.2012.6335244 

Ghanbari, M., Najafi, G., Ghobadian, B., Mamat, R., Noor, M. M., & Moosavian, A. (2015). 
Adaptive neuro-fuzzy inference system (ANFIS) to predict CI engine parameters 
fueled with nano-particles additive to diesel fuel. In IOP Conference Series: Materials 
Science and Engineering, Kuantan, Pahang, Malaysia, August 18-19, Volume 100, No. 
1 (pp. 1-8). IOP Publishing 

Gokdag, M., Hasiloglu, A. S., Karsli, N., Atalay, A., & Akbas, A. (2007). Modeling of 
vehicle delays at signalized intersection with an adaptive neuro-fuzzy (ANFIS). 
Journal of Scientific and Industrial Research, 66(9), 736-740. 

Hasiloglu, A. S., Gokdag, M., & Karsli, N. (2014). Comparison an artificial intelligence-
based model and other models: Signalized intersection delay estimates. International 
Journal of Engineering and Innovative Technology, 4(3), 220-231. 

Hosoz, M., Ertunc, H. M., Karabektas, M., & Ergen, G. (2013). ANFIS modelling of the 
performance and emissions of a diesel engine using diesel fuel and biodiesel blends. 
Applied Thermal Engineering 60(1-2), 24-32. 
https://doi.org/10.1016/j.applthermaleng.2013.06.040 

Hosseinlou, M. H., & Sohrabi, M. (2009). Predicting and identifying traffic hot spots 
applying neuro-fuzzy systems in intercity roads. International Journal of 
Environmental Science & Technology, 6(2), 309-314. 
https://doi.org/10.1007/BF03327634 

Hou, Y., Zhang, J., Zhang, Y., & Chen, L. (2008). Integrated chassis control using ANFIS. 
In Proceedings of IEEE International Conference on Automation and Logistics, 
Qingdao, China, September 1-3, (pp. 1625-1630). IEEE 10.1109/ICAL.2008.4636414 

Jomaa, D., Yella, S., & Dougherty, M. (2015). Automatic Trigger Speed for Vehicle 
Activated Signs using Adaptive Neuro fuzzy system and Classification Regression 



Stojčić/Oper. Res. Eng. Sci. Theor. Appl. 1 (1) (2018) 40-61 

 

58 
 

Trees. In Fourth International Conference on Intelligent Systems and Applications 
INTELLI 2015, St. Julians, Malta, October 11-16 (138). 

Joshi, M., & Hadi, T. H. (2015). A review of network traffic analysis and prediction 
techniques. arXiv preprint arXiv:1507.05722.  

Jang, J. S. (1993). ANFIS: adaptive-network-based fuzzy inference system. IEEE 
transactions on systems, man, and cybernetics, 23(3), 665-685. 10.1109/21.256541 

Jang, J. S. (1996). Input selection for ANFIS learning. In Proceedings of IEEE 5th 
International Fuzzy Systems, New Orleans, LA, USA, September 11, Volume 2 (pp. 
1493-1499). IEEE 10.1109/FUZZY.1996.55239 

Kar, S., Das, S., & Ghosh, P. K. (2014). Applications of neuro fuzzy systems: A brief 
review and future outline. Applied Soft Computing, 15, 243-259. 
https://doi.org/10.1016/j.asoc.2013.10.014 

Khodayari, A., Kazemi, R., Ghaffari, A., & Manavizadeh, N. (2010). Modeling and 
intelligent control design of car following behavior in real traffic flow. Proceedings of 
IEEE Conference on Cybernetics and Intelligent Systems, Singapore, June 28-30 (pp. 
261-266). IEEE 10.1109/ICCIS.2010.5518546 

Korkmaz, E. & Akgüngör A. P. (2016). An application of adaptive neuro-fuzzy 
inference system in vehicle delay modeling. International Journal of Advances in 
Mechanical and Civil Engineering, 3(4), 59-62. IJAMCE-IRAJ-DOI-5393 

Kothandaraman, R., & Ponnusamy, L. (2012). PSO tuned adaptive neuro-fuzzy 
controller for vehicle suspension systems. Journal of advances in information 
technology, 3(1), 57-63. 10.4304/jait.3.1.57-63 

Kumar, M. K., & Prasad, V. K. (2015). Driver Behavior Analysis and Prediction Models: 
A Survey. International Journal of Computer Science and Information 
Technologies, 6(4), 3328-3333. 

Kukadapwar, S. R., & Parbat, D. K. (2015). Modeling of traffic congestion on urban 
road network using fuzzy inference system. American Journal of Engineering 
Research, 4(12), 143-148. 

Lai, G. R., Soh, A., Sarkan, H. M. D., Rahman, R. A., & Hassan, M. K. (2015). Controlling 
traffic flow in multilane-isolated intersection using ANFIS approach techniques. 
Journal of Engineering Science and Technology, 10(8), 1009-1034. 

Li, X. X., Huang, H., & Liu, C. H. (2007). The application of an ANFIS and BP neural 
network method in vehicle shift decision. Paper presented at the 12th World 
Congress in Mechanism and Machine Science, Besançon, France, June 18-21. 

Liu, M., & Chen, Y. (2017). Predicting Real-Time Crash Risk for Urban Expressways in 
China. Mathematical Problems in Engineering, 2017, 1-10. 
https://doi.org/10.1155/2017/6263726 

Massoud, M., Sahebgharani, A. & Ramesht, Z. (2014). Eveloping an Integrated Land 
Use-Transportation Model (Ilutm) For Analysis and Prediction of Energy 
Consumption Through Adaptive Neuro-Fuzzy Inference System (ANFIS). 
International Journal of Recent Scientific Research, 5(1), 107-109.  

Maurya, A. K., & Patel, D. K. (2015). Vehicle Classification Using Adaptive Neuro-Fuzzy 
Inference System (ANFIS). In K. N. Das, D. Kusum, M. Pant, B. C. Jagdish, N. Atulya 
(Eds.), Proceedings of Fourth International Conference on Soft Computing for 
Problem Solving (pp. 137-152). Springer India 



Application of ANFIS model in road traffic and transportation: a literature review from 1993 to 
2018 

 

59 
 

 

Maghsoudi, R., & Moshiri, B. (2017). Applying Adaptive Network-Based Fuzzy 
Inference System to Predict Travel Time in Highways for Intelligent Transportation 
Systems. Journal of Advances in computer Research, 8(3), 87-103. 

Mohebbi, M., Charkhgard, M., & Farrokhi, M. (2005). Optimal neuro-fuzzy control of 
parallel hybrid electric vehicles. In Proceedings of IEEE Vehicle Power and Propulsion 
Conference, Chicago, IL, USA, September 7 (pp. 26-30). IEEE 
10.1109/VPPC.2005.1554566 

Nugroho, P. W., Li, W., Du, H., Alici, G., & Yang, J. (2014). An adaptive neuro fuzzy 
hybrid control strategy for a semiactive suspension with magneto rheological 
damper. Advances in Mechanical Engineering, 6, 487312. 
https://doi.org/10.1155/2014/487312 

Ottomanelli, M., Caggiani, L., Iannucci, G., & Sassanelli, D. (2010). An adaptive neuro-
fuzzy inference system for simulation of pedestrians behaviour at unsignalized 
roadway crossings. In Gao XZ., Gaspar-Cunha A., Köppen M., Schaefer G., Wang J. 
(Eds.), Soft computing in industrial applications (pp. 255-262). Springer, Berlin, 
Heidelberg https://doi.org/10.1007/978-3-642-11282-9_27 

Pamučar, D., Bozanic, D., & Komazec, N. (2016). Managing the hazardous material 
transportation process using the adaptive neural networks and Dijkstra algorithm. In 
Proceedings of II International Scientific Conference Safety and Crisis Management–
Theory and Practise Safety for the Future, Obrenovac, Serbia (pp. 112-116). 

Pamučar, D., Ljubojević, S., Kostadinović, D., & Đorović, B. (2016a). Cost and risk 
aggregation in multi-objective route planning for hazardous materials 
transportation—A neuro-fuzzy and artificial bee colony approach. Expert Systems 
with Applications, 65, 1-15. https://doi.org/10.1016/j.eswa.2016.08.024 

Pamučar, D., & Ćirović, G. (2018). Vehicle route selection with an adaptive neuro 
fuzzy inference system in uncertainty conditions. Decision Making: Applications in 
Management and Engineering, 1(1), 13-37. doi.org/10.31181/dmame180113p 

Pamučar, D., Lukovac, V., & Pejčić-Tarle, S. (2013). Application of Adaptive Neuro 
Fuzzy Inference System in the process of transportation support. Asia-Pacific Journal 
of Operational Research, 30(02), 1250053. 
https://doi.org/10.1142/S0217595912500534 

Pérez, J., Gajate, A., Milanés, V., Onieva, E., & Santos, M. (2010). Design and 
implementation of a neuro-fuzzy system for longitudinal control of autonomous 
vehicles. In Proceedings of International Conference on Fuzzy Systems, Barcelona, 
Spain, July 18-23 (pp. 1-6). IEEE 10.1109/FUZZY.2010.5584208 

Poor Arab Moghadam, M., Pahlavani, P., & Naseralavi, S. (2016). Prediction of car 
following behavior based on the instantaneous reaction time using an ANFIS-CART 
based model. International Journal of Transportation Engineering, 4(2), 109-126. 
10.22119/IJTE.2016.40536 

Rai, A. A., Pai, P. S., & Rao, B. S. (2015). Prediction models for performance and 
emissions of a dual fuel CI engine using ANFIS. Sadhana, 40(2), 515-535. 
https://doi.org/10.1007/s12046-014-0320-z 

Rahimi, A. M. (2017). Neuro-fuzzy system modelling for the effects of intelligent 
transportation on road accident fatalities. Tehnički vjesnik, 24(4), 1165-1171. 
https://doi.org/10.17559/TV-20151203230220 



Stojčić/Oper. Res. Eng. Sci. Theor. Appl. 1 (1) (2018) 40-61 

 

60 
 

Saadeddin, K., Abdel-Hafez, M. F., Jaradat, M. A., & Jarrah, M. A. (2013). Performance 
enhancement of low-cost, high-accuracy, state estimation for vehicle collision 
prevention system using ANFIS. Mechanical Systems and Signal Processing, 41(1-2), 
239-253. https://doi.org/10.1016/j.ymssp.2013.06.013 

Saifizul, A. A., Zainon, M. Z., & Osman, N. A. (2006). An ANFIS controller for vision-
based lateral vehicle control system. In Proceedings of 9th International Conference 
on Control, Automation, Robotics and Vision, Singapore, December 5-8 (pp. 1-4). IEEE 
10.1109/ICARCV.2006.345277 

Seesara, S. & Gadit, J. (2015). Smart Traffic Control Using Adaptive Neuro-Fuzzy 
Inference System(ANFIS). International Journal of Advance Engineering and Research 
Development, 2(5), 295-302. DOI:10.21090/IJAERD.020540 

Selma, B., & Chouraqui, S. (2012). Trajectory estimation and control of vehicle Using 
neuro-fuzzy technique. International Journal of Advances in Engineering & 
Technology, 3(2), 97. 

Sharma, A., Vijay, R., Bodhe, G. L., & Malik, L. G. (2014). Adoptive neuro-fuzzy 
inference system for traffic noise prediction. International Journal of Computer 
Applications, 98(13), 14-19. 

Shankar, H., Raju, P. L. N., & Rao, K. R. M. (2012). Multi model criteria for the 
estimation of road traffic congestion from traffic flow information based on fuzzy 
logic. Journal of Transportation Technologies, 2(01), 50. 10.4236/jtts.2012.21006 

Shuliakov, V., Nikonov, O., & Fastovec, V. (2015). Application of Adaptive Neuro-Fuzzy 
Regulators in the Controlled System by the Vehicle Suspension. International Journal 
of Automation, Control and Intelligent Systems, 1(3), 66-72. 

Sindhuja, V., & Ranjitham, G. (2014). Regenerative Braking System of Electric Vehicle 
Driven by BLDC Motor Using Neuro Fuzzy and PID. International Journal of 
Innovative Research in Science, Engineering and Technology, 3(12), 17847-17854.  

Sun, R., Xie, F., Xue, D., Zhang, Y., & Ochieng, W. Y. (2017). A novel rear-end collision 
detection algorithm based on GNSS fusion and ANFIS. Journal of Advanced 
Transportation, 2017, 1-10. https://doi.org/10.1155/2017/9620831 
Syahputra, R. (2016). Application of Neuro-Fuzzy Method for Prediction of Vehicle 
Fuel Consumption. Journal of Theoretical & Applied Information Technology, 86(1), 
138-150. 

Thipparat, T., & Thaseepetch, T. (2012). Application of Neuro-Fuzzy System to 
Evaluate Sustainability in Highway Design. International Journal of Modern 
Engineering Research, 2(5), 4153-4158. 

Udofia, K. M., Emagbetere, J. O., & Edeko, F. O. (2014). Dynamic traffic signal phase 
sequencing for an isolated intersection using ANFIS. Automation, Control and 
Intelligent Systems, 2(2), 21-26. 10.11648/j.acis.20140202.12 

Wannige, C. T., & Sonnadara, D. U. J. (2008). Traffic signal control based on adaptive 
neuro-fuzzy inference. In A. Punchihewa, B. Dias (Eds.), Proceedings of 4th 
International Conference on Information and Automation for Sustainability, Colombo, 
Sri Lanka, December 12-14 (pp. 301-306). IEEE 10.1109/ICIAFS.2008.4783976 

Wankhede, S. S., Khanapurkar, M. M., & Bajaj, P. (2011). Intelligent Speed Adaptation 
System with Hybrid Algorithm. International Journal of Computer Science and 
Network Security, 11(1), 97. 



Application of ANFIS model in road traffic and transportation: a literature review from 1993 to 
2018 

 

61 
 

 

Zaki, J. F., Ali-Eldin, A. M. T., Hussein, S. E., Saraya, S. F., & Areed, F. F. (2016). 
Framework for Traffic Congestion Prediction. International Journal of Scientific & 
Engineering Research, 7(5), 1205-1210. 

Özkan, İ. A., Ciniviz, M., & Candan, F. (2015). Estimating Engine Performance and 
Emission Values Using ANFIS/ANFIS Kullanılarak Motor Performans ve Emisyon 
Değerleri Tahmini. International Journal of Automotive Engineering and 
Technologies, 4(1), 63-67. http://dx.doi.org/10.18245/ijaet.95440