alexa An Integrated Model to Control Traffic Lights Controlling of Traffic Lights in Multiple Intersections Using Fuzzy Logic and Genetic Algorithm | Open Access Journals
ISSN: 2151-6219
Business and Economics Journal
Make the best use of Scientific Research and information from our 700+ peer reviewed, Open Access Journals that operates with the help of 50,000+ Editorial Board Members and esteemed reviewers and 1000+ Scientific associations in Medical, Clinical, Pharmaceutical, Engineering, Technology and Management Fields.
Meet Inspiring Speakers and Experts at our 3000+ Global Conferenceseries Events with over 600+ Conferences, 1200+ Symposiums and 1200+ Workshops on
Medical, Pharma, Engineering, Science, Technology and Business

An Integrated Model to Control Traffic Lights Controlling of Traffic Lights in Multiple Intersections Using Fuzzy Logic and Genetic Algorithm

Khaled Abdul Rahman Jomaa* and Cheng Jack Kie

Faculty of Industrial Management, University Malaysia Pahang, Kuantan-Pahang-Malaysia

*Corresponding Author:
Jomma KAR
Faculty of Industrial Management
University Malaysia Pahang
Kuantan-Pahang-Malaysia
Tel: +60 9-424 5000
E-mail: [email protected]

Received date: January 16, 2017; Accepted date: January 24, 2017; Published date: January 31, 2017

Citation: Jomaa KAR, Kie CJ (2017) An Integrated Model to Control Traffic Lights Controlling of Traffic Lights in Multiple Intersections Using Fuzzy Logic and Genetic Algorithm. Bus Eco J 8:287. doi: 10.4172/2151-6219.1000287

Copyright: © 2017 Jomaa KAR, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Visit for more related articles at Business and Economics Journal

Abstract

In this paper we propose an integrated model combines Fuzzy Logic (FL) and Genetic Algorithm (GA), utilizing their applications in order to minimize the traffic congestion and traffic delay, through controlling traffic light system in three proposed traffic intersections. The proposed model in this paper will adjust the timing and phasing of the green traffic lights according to the current situation in the proposed traffic intersections; each intersection is supposed to be controlled by traffic signals that will apply the model. The green light interval time length shall provide at an intersection will be decided by FL. the outputs of FL will be optimized by GA, in order to obtain a higher performance. This performance can be measured considering the reduction in the waiting time and the total amount of vehicles that arrived to the Queue of the three intersections. The proposed model expected to provide a significant improvement to the traffic light system performance which might be very important to be applied in the metropolitan areas in Malaysia.

Keywords

Traffic Light; Fuzzy Logic; Genetic Algorithm; Congestion; Malaysia

Introduction

Many developing Asian cities are finding it is difficult to cope with the growth of urbanization and increased car ownership. Traffic congestion is a growing problem in many metropolitan areas in Malaysia. Congestion increases travel time, air pollution, carbon dioxide (CO2) emissions and fuel use because cars cannot run efficiently. Studies by economists at the World Bank have shown that in 2014, economic losses due to traffic congestion in Greater Kuala Lumpur was estimated at 1.1-2.2% of GDP in 2014 [1]. The majority of this cost is associated with lost productivity, followed by wasted fuel costs and environmental damage caused by the tailpipe exhaust fumes from wasted fuel.

One of the best ways to eliminate traffic congestion is to control traffic flow through controlling traffic lights, in order to eliminate the conflicting traffic flows between the consecutive traffic intersections, and using intelligent traffic light system, Since an intelligent traffic light system is a major part of an intelligent transportation system, it is a challenge to increase the work efficiency of these traffic signals in order to reduce traffic jams and congestions as well as vehicle emissions in the metropolitan areas.

The current system of traffic light in many traffic intersection in Malaysia still using a conventional system which have been providing a fixed traffic control plan, this system is called “Pre-set Cycle Time (PCT) Controller” the setting for this kind of traffic lights is almost be done based on prior traffic counts, and mostly the setting can be manually changed, the disadvantage of this method that if there is a congestion in any roadway, the green light will not be extended, and the next phase will continue on time without considering the traffic density in any other junction [2].

Another common conventional traffic light system nowadays called Vehicle Actuated (VA) Controllers, in this method every road is built with only one magnetic sensor placed at the road end before the junction. The sensor is usually visible as a black rectangular line next to the traffic light in each road. This method works with three parameters: Initial Interval, Extension Unit and Extension Limit [3], if the phase turned to green light, the green phase will be expended by the extension unit while there are a cars crossing the junction, ones the extension limit is reached the traffic light will turned to red phase. This method also doesn’t consider the traffic density in any other junctions, and doesn’t give priority to the most congested road.

As the number of vehicles in Malaysia is increasing while the road system doesn’t be improved well, therefore, the current traffic light system is not capable of solving the problem of congestion by using the same efficiency. Therefore it is very necessary to apply an intelligent traffic light system that overcomes the congestion and traffic jam. The traffic light system should be able to interact with congestion and thus help passengers, drivers, and products to reach their destinations quickly and safely. Which might save time, effort and, minimize the transportation costs.

In this paper we propose an integrated model including a genetic algorithm (GA), and Fuzzy Logic (FL); in order to adjust the timing and the green phase of traffic light according to the current situation in multiple intersections; every intersection will be controlled by traffic signals. Thus, the model will decide how long the green light interval time shall be provided in each road at an intersection; in order to decrease the traffic congestion.

Statement of the Problem

The rapid population growth in Malaysia with depending on private cars for transportation has its impact on traffic congestion in the metropolitan areas; in the shade of using the current conversional traffic light system the traffic congestion problem get worst. The big challenge we face now is how to minimize the traffic congestion in the road system and to reach to smooth and safe traffic flow.

In order to elicit a fundamental diagnosis about traffic congestion problem in Malaysia, a cluster random sample has been selected using Google forms technique. In this regard a survey has been conducted on 1360 respondents in Malaysia [4], the survey has revealed that 50.5% of the respondents in KL reported that the main problem in the traffic system that causes the high level of traffic congestion is “Inadequate green time in traffic lights” (Table 1), while 22.1% of them reported that traffic accidents in the road way what cause the high level of traffic congestion

The most critical Transportation Problem Kuala Lumper Other Grand Total
Frequency Percentage Frequency Percentage Frequency Percentage
Inadequate green time in traffic lights 301 50.50% 395 51.70% 696 51.20%
Road utility work 64 10.70% 80 10.50% 144 10.60%
The attitude of the driver on the road 99 16.60% 135 17.70% 234 17.20%
Traffic incidents 132 22.10% 154 20.20% 286 21.00%
Grand Total 596 100.00% 764 100.00% 1360 100.00%

Table 1: The most critical causes of traffic congestion problem in Malaysia.

These findings may support our idea to find the best way to manage the time in the traffic light system in order to minimize the waiting time.

It is clear from (Table 2) that the majority of the population in Malaysia depends on privet cars as the main transportation method, where 53.6% of total respondents and 50.5% of those who live in KL revealed that they usually use private care for transportation, while only 29.5% of the respondents in KL reported that they use train for transportation.

Kind of Vehicle Used Kuala Lumper Other Grand Total
Frequency Percentage Frequency Percentage Frequency Percentage
Motorbike 52 8.70% 67 8.80% 119 8.80%
Private Car 301 50.50% 428 56.00% 729 53.60%
Public Bus 33 5.50% 42 5.50% 75 5.50%
Taxi 34 5.70% 31 4.10% 65 4.80%
Train 176 29.50% 196 25.70% 372 27.40%
Grand Total 596 100.00% 764 100.00% 1360 100.00%

Table 2: The most common transportation method used.

In this sense the big number of private car users may minimize the capacity of the road system; this may also increase the level of congestion. So that this is very important to find good methods to manage the traffic flow density and to minimize the waiting time in the traffic lights.

The aim of this study is to develop an integrated model to control traffic light System as an application for the management of traffic congestion which might be good method to be applied in the metropolitan areas in Malaysia.

Objectives

The objectives of this research are to introduce a proposed integrated model to achieve the following objectives:

1. Providing an efficient method for traffic flow management.

2. Organizing and optimize the green time interval for traffic light system.

3. Smoothing the traffic flow and eliminating the conflict between multiple consecutive traffic intersections.

4. Minimizing the traffic congestion in multiple consecutive intersections.

Scope of the Work

Kuala Lumpur city is selected as a study area for assessing the traffic congestion. Kuala Lumpur is considered as the most affected area by traffic congestion (Figures 1-3) is depicting the road net-work of Kuala Lumpur, which provides data about real time road congestion right on the map. Red color shows block way, yellow color shows slow moving, and green colour shows smooth flowing.

business-economics-net-work

Figure 1: The road net-work in Kuala Lumpur.

business-economics-net-work

Figure 2: The typical intersection with Two Bunches of traffic Phases.

business-economics-logic-controllers

Figure 3: Structure of fuzzy Logic Controllers FL.

We will construct our model and assess the results on simulated data, and then the parameters of the model will be revised to adopt real situation in KL.

Review of Related Work

Many researchers recently have focused on finding a new methods to control signals that adjust the timing and phasing of lights in order to minimize traffic congestion, some of these studies used one method like FL or GA, and few studies used hybrid model including these two methods; some studies focus on an isolated traffic intersection, and others focus on multiple intersection. Here we present some of these studies.

A two stage traffic light system for real-time traffic monitoring has been proposed by Alam [5] to dynamically manage both the phase and green time of traffic lights for an isolated signalized intersection with the objective of minimizing the average vehicle delay in different traffic flow rate. There are two different modules namely traffic urgency decision module (TUDM) and extension time decision module (ETDM). A mathematical model representing the traffic control stochastic environment has been developed by Ezzat et al., [6]. The optimum/ near optimum traffic signal timing values have been determined through the application of a genetic algorithm that feeds these values into a developed simulation model to obtain the corresponding queuing parameters. The generated signal timings significantly enhance the traffic performance and alleviate the choke points over a multiple-junction urban network. Optimization results show that the proposed model can improve the parameters of the vehicular flow. The chromosomes which represents the cycle time and the green split, is generated randomly in this model which might include some mistakes. A fuzzy logic for multi-agent based autonomous traffic lights control system was applied for to intersections by Salehi et al., [7]. The real time parameters such as traffic density and queue length are obtained by using image-processing techniques. Singh et al., [8] used a genetic algorithm and traffic emulator, developed in JAVA, to represent dynamic traffic conditions the high proportion of motorcycles. Turky et al., [9] proposed an optimal control of traffic lights using genetic algorithm (GA), in a four-way, two-lane junction with a pedestrian crossing to provide intelligent green interval responses based on dynamic traffic load inputs, they compared the performance between the genetic algorithms controller and a conventional fixed time controller and the results showed that the genetic algorithms controller performs better than the fixed-time controller.

Methodology

Introduction

This study propose an integrated model including a genetic algorithm (GA), and Fuzzy Logic (FL); in order to adjust the timing and the green phase of traffic light according to the current situation in multiple intersections; the data about traffic density will be captured and collected using a placed observing camera to be used in our model. Every intersection will be controlled by traffic signals. Thus, the model will decide how long the green light interval time shall be provided in each road at an intersection; in order to decrease the traffic congestion. The proposed model expected to provide a good contribution to the current knowledge in this field.

In this study, both analytical and quantitative approaches are utilized. The analytical approach is utilized to analyze the performance of the Malaysian traffic system. The quantitative approach is used within an integrated transportation model framework utilizing Genetic Algorithm (GA) and Fuzzy Logic (FL) Methods.

Here we present the research methodology that was developed and followed in conducting this paper. We present also the methods that we use to construct our model, how we structured the model. We also present the model structure and the methods that we follow to collect the data.

Traffic flow phases

Our proposed model will be applied for two bunches of traffic phases as it shown in Figures 2 and 4 in order to decide which bunch of traffic phases should be applied for the model to give optimal results.

business-economics-flow-chart

Figure 4: Flow chart of Model.

Video image detection systems

The objects in the road will be defined by the Video image detection system as vehicles moving or stopping on the roads. The different sizes of the vehicles that use the road can be observed and counted. The basic algorithm starts with a pre-image processing step, consisting of digitization and segmentation. The next step is called video segmentation which can be defined as the following: given a scene of the road without any object captures this frame as a reference frame, which will be considered as background to our video segmentations that may contain one or more objects. The algorithm has four major functions, the first one is convert the video segmentation into frames and select the last frame and will call it current image, the second function is convert the current image to black and white in order to prepare for boundary tracing and remove the noise by using morphology functions. The third function subtracts the background from the current frame. The last function counts all distinct objects in the image [1].

The model will be applied to manage three intersections, A, B, and C, with distance 200 m between each two intersections; each intersection is controlled by a smart traffic light, each intersection consists of four roads, and each road consists of three directions (Straight, Left, and Right). The model is assumed to be used in the most congested areas so we assume that for all directions in all intersections there should be a number of vehicles that will travel in the same speed 40 km/h.

Fuzzy logic

In this proposed model, the input data of the number of vehicles on each lane of the road in each intersection will be taken from a video object detection system. The data will be treated by fuzzy logic that is characterized by a set of rules which defined antecedent.

There are two inputs representing the number of vehicles in the same lane and the number of vehicles in the affected lane. For example, the number of vehicles at lane 2 on intersection B Figure 1 will affect the number of vehicles on lane 4 in intersection A. However, in other cases, there will be three inputs because of the existence of two affected lanes, such as (C1) the number of vehicles at lane 1 of the intersection C which will have an effect on (B1) the number of vehicles at lane 1 in intersection B and also (A4) the number of vehicles at lane 4 in intersection A. In this case the fuzzy logic system will work in two stages: Fuzzy Logic Controller FLC1 will be done for the first two inputs; then the output of this (FLC1) will be considered as an input for the second stage of FLC2 including the third input.

Figure 3 shows the structure of the system composed of FLC1 and FLC2; FLC1 has two inputs that represent C1 and B1, respectively, and FLC2 has also two inputs In (A4) and the output of FLC1 that represent the green light time interval for the signal at C1. The output of FLC2 represents the final decision of green light time interval for a signal B1.

Fuzzy logic rules include all possible scenarios for each traffic jam in every intersection; finally we can get two outputs from fuzzy logic model, the cycle time and green light time for each lane in each intersection. Then GA is applied to optimize the best performance of FLC.

Genetic Algorithm GA

Each chromosome is composed of (n) genes representing (i+k) decision variables. These are the green splits and the cycle times of the junctions in the road network which have been generated in fuzzy logic model. The chromosome size depends on the number of feasible states (i) and the number of intersections in the road network (k) which we assume three as three intersections in our paper. The fitness function of each chromosome, the same as the objective function in the mathematical model (Equation 1), is evaluated. The chromosomes are then ranked according to their normalized fitness values. As stated earlier, the chromosome size depends on the size of the road network under study. then the model check, if termination criteria is satisfied, which means either the predefined maximum number of generation have reached or fitness function is not satisfied, If the termination criteria are not satisfied; selection is performed from the given population to obtain fitter parents, which can lead to fitter sons. These parents, thus selected are mated to produce fitter children and this phenomenon is called crossover or recombination. Some mutation is performed for example some bits of children are altered from the above result. After mutation we have a new set of generation, now go back to evaluating the fitness function. If the termination criteria are satisfied, get the solution.

The result is received from GA is the green extension times for all roads. These extension times are added with predefined fixed green times and applied to the traffic light system.

Mathematical model

We use the structure of Turky et al., [9 ] model. The characteristic of objective function, as shown in Equation 1, is non-linear function which includes two performance metrics: Queue Length and vehicular waiting time. Queue Length and Waiting time Equations 5 and 6 compute the maximum values of Queue Length and Waiting time occurring for each control point (j) in junction (k) at period (p). The maximum value in this array is considered as the objective function to be minimized.

The decision variables are the green splits of each state (i) and the cycle time of each junction (k) in the network at different periods, denoted by which represent the proportion of state (i) from the total cycle time of the junction (k), and which can be counted as the sum of all the state timings occurring at that junction, which also must equal the green and red signal timings for control point (j) in the network, as shown in Equation 4. Equations 7 and 8 determine the average of the maximum queue length and waiting time over a specified period (p); the rest model equations 9-11 are set of hard constraints.

Objective Minimize:

Z

LQmax jkp Maximum Vehicular Queue Length for control point (j) in junction (k) at period (p)

WQmax jkp Maximum Vehicular waiting time for control point (j) in junction (k) at period (p)

Subject to

Tikp State timing interval of state (i) in junction (k) at period (p)

gikp Green Split of state (i) in junction (k) at period (p)

CTkp Cycle Time of junction (k) at period (p)

CTkp Cycle Time of junction (k) at period (p)

CTkp Cycle Time of junction (k) at period (p)

Sijkp Current Signalization Setting (Green/Red) within a state (i) for a control point (j) in junction (k) at period (p)

LQmax jkp Maximum Vehicular Queue Length for control point (j) in junction (k) at period (p)

Ajkp Dynamic Arrival inflow at control point (j) in junction (k) at period (p)

Djkp Dynamic Dissipation outflow at control point (j) in junction (k) at period (p)

WQmax jkp Maximum Vehicular waiting time for control point (j) in junction (k) at period (p)

TAvjkp Arrival time of vehicle (v) at control point (j) in junction (k) at period (p)

TDvjkp Departure time of vehicle (v) at control point (j) in junction (k) at period (p)

Average of Maximum queue lengths for a period (p)

j Control Point index, where

k Junction index, where

p Period index, where

Average of Maximum vehicular waiting times for a period (p)

Tmin Lower bounds for a state timing.

Tikp State timing interval of state (i) in junction (k) at period (p)

Tmax Upper bounds for a state timing.

CTmin Lower bounds for a junction’s cycle time.

CTkp Cycle Time of junction (k) at period (p)

CTmax Upper bounds for a junction’s cycle time.

gikp Green Split of state (i) in junction (k) at period (p)

Expected Results

We intend to expand our proposed model, so Neural Network NN method, and simulation model will be added in order to make it more robust; The validity of the model will be checked at the first using simulation model that it will be applied in MATLAB code; finally we can apply our model in Kuala Lumpur city, using the real time data using the observing camera and strategically placed ultra-sonic sensors.

With applying and testing this proposed model we expect to get the following outcomes:

- Optimal traffic lights’ timing control at the targeted traffic intersections.

- Eliminated conflict between multiple consecutive traffic intersections

- Eliminated Traffic congestion with Smooth traffic flow of vehicles.

- Applying an accurate model for traffic is close to reality, using a real time data captured by the observing camera system; which might be adjustable to be applied in Malaysian traffic light system.

References

Select your language of interest to view the total content in your interested language
Post your comment

Share This Article

Relevant Topics

Article Usage

  • Total views: 659
  • [From(publication date):
    March-2017 - Nov 24, 2017]
  • Breakdown by view type
  • HTML page views : 570
  • PDF downloads : 89
 

Post your comment

captcha   Reload  Can't read the image? click here to refresh

Peer Reviewed Journals
 
Make the best use of Scientific Research and information from our 700 + peer reviewed, Open Access Journals
International Conferences 2017-18
 
Meet Inspiring Speakers and Experts at our 3000+ Global Annual Meetings

Contact Us

Agri & Aquaculture Journals

Dr. Krish

[email protected]

1-702-714-7001Extn: 9040

Biochemistry Journals

Datta A

[email protected]

1-702-714-7001Extn: 9037

Business & Management Journals

Ronald

[email protected]

1-702-714-7001Extn: 9042

Chemistry Journals

Gabriel Shaw

[email protected]

1-702-714-7001Extn: 9040

Clinical Journals

Datta A

[email protected]

1-702-714-7001Extn: 9037

Engineering Journals

James Franklin

[email protected]

1-702-714-7001Extn: 9042

Food & Nutrition Journals

Katie Wilson

[email protected]

1-702-714-7001Extn: 9042

General Science

Andrea Jason

[email protected]

1-702-714-7001Extn: 9043

Genetics & Molecular Biology Journals

Anna Melissa

[email protected]

1-702-714-7001Extn: 9006

Immunology & Microbiology Journals

David Gorantl

[email protected]

1-702-714-7001Extn: 9014

Materials Science Journals

Rachle Green

[email protected]

1-702-714-7001Extn: 9039

Nursing & Health Care Journals

Stephanie Skinner

[email protected]

1-702-714-7001Extn: 9039

Medical Journals

Nimmi Anna

[email protected]

1-702-714-7001Extn: 9038

Neuroscience & Psychology Journals

Nathan T

[email protected]

1-702-714-7001Extn: 9041

Pharmaceutical Sciences Journals

Ann Jose

[email protected]

1-702-714-7001Extn: 9007

Social & Political Science Journals

Steve Harry

[email protected]

1-702-714-7001Extn: 9042

 
© 2008- 2017 OMICS International - Open Access Publisher. Best viewed in Mozilla Firefox | Google Chrome | Above IE 7.0 version
adwords