1. Introduction
Traffic congestion is becoming a major and common problem in more and more countries, and especially in developed cities with an increased population density, the number of cars is automatically high. This has a negative impact on a region’s economy and social development. In the case of those who travel, traffic jams not only increase the time and costs of the trip but also the quality. From the point of view of traffic management, the results of road congestion increase operating costs and decrease the efficiency of traffic networks. The chain effect caused by traffic congestion also affects the social factor, causing air pollution, traffic accidents, noise pollution, and also other problems that degrade the environment. With over 1.3 million victims and about 50 million people injured annually in road accidents [
1], this problem definitely belongs to our society. The economic cost of these road accidents is estimated at between 1% and 3% of a country’s gross domestic product [
2]. According to existing estimates, as a percentage, approximately 90% of road accidents occur due to various causes generated by human error, lack of attention, fatigue, delayed reactions, or lack of distributive attention [
1,
2]. Current active safety systems have a proven potential to significantly reduce accidents. Such systems are based on sensors that identify and highlight a situation with a potential danger, reducing the danger or avoiding an accident [
3].
The analysis carried out in the field of inter-vehicle communications (V2V) and infrastructure-vehicle communications (I2V) has the potential to reduce by up to 80% of dangerous traffic situations, taking vehicle safety to another level. Therefore, communication-based road safety systems allow an infrastructure unit or vehicle to be informed of a high-risk traffic situation by informing the driver and other road users of that situation [
3]. Thus, the vehicles that are part of that network can change information on speed and location, but also maintain a constant speed, accelerating and decelerating according to the other vehicles in the network, with their cooperation being useful both for the prevention of traffic situations that can generate accidents, but also for the increase of traffic fluidity [
4].
However, supporting the safety of communications-based vehicles is difficult in light of strict requirements and standards. These applications require fairly low latencies that can reach up to 20 ms to detect collisions and pre-crash, so the literature shows that packet data delivery and standard communication varies up to 300 m [
4]. Research has been conducted on the formation and spread of traffic congestion, being a rather pressing issue of today’s society and causing quite serious problems. The causes of traffic congestion can be listed in three categories: (1) temporary barrier control, (2) network blockage in key areas, and (3) random fluctuation of a region in the network. All known or unknown causes and events in traffic affect the flow and influence the smooth running of things, and they can be collectively referred to as traffic accidents [
5,
6].
We can say that a large part of traffic accidents cause congestion and congestion causes traffic accidents. Road traffic works like a network, if the accident or event is remedied in time, the vehicle is put back on track and continues on its way; otherwise, traffic congestion will spread quickly in the network. Studies confirm that exiting the column and leaving the direction of travel is unreasonable, forming loop congestion, a phenomenon that does not dissipate congestion [
6]. Forming a closed loop congestion adjusts the time and blockage, even to the starting point of the network. We can say that as long as a closed loop is formed, the traffic congestion itself becomes difficult to dissipate, especially without artificial interventions.
In this paper, simulations are made on the propagation of traffic and the occurrence of congestion depending on the density of cars and the established graphic model, thus obtaining defining elements through which traffic congestion can be dispelled using means to streamline traffic, whether we are talking about intelligent traffic systems, autonomous cars, or dedicated applications. Most traffic events are caused by road accidents or various activities undertaken by road users or even pedestrians, and each delay in starting a car produces a knot in the car network, and from that point to a congestion becomes a matter of a few minutes or even seconds. The simulations performed are based on relevant traffic data, both quantitative and qualitative on the density of cars, but also the hours with an increased flow of vehicles.
Models designed for simulation are based on learning and calibrating parameters, optimizing macro-model, macro-network, node, and intersection simulations [
7]. Looking in detail at this research, which is based on detailed analyses of the state of traffic, forecasts for urban road traffic can be made and improved by providing technical assistance solutions in order to establish the impact on the flow of cars transiting a certain area. This direction of research offers development perspectives, being a field of great interest, especially since with the increased living standard of the population, automatically, the number of vehicles has increased exponentially, whereas most road infrastructures have remained the same. The main purpose is to efficiently manage the information and to highlight possible solutions in solving the mentioned problem, whether we are talking about autonomous vehicles, vehicle-to-vehicle (V2V), vehicle-to-infrastructure (V2I), vehicle-to-everything (V2X) communications [
8,
9], or visible light communication (VLC) [
10,
11]. All these elements approached in this direction will have a beneficial contribution to the previously mentioned problems, becoming a starting point for the fields dedicated to intelligent traffic systems and safe driving, but especially the reduction of the number of victims of road events, which is a global problem that produces millions of victims annually.
4. Control Strategy and Potential Solution
4.1. Introduction to Control Strategy
For a better representation, vehicle generation schemes were designed for each direction of travel, each receiving a synchronized value depending on the desired scenario. Thus, in a first phase, the density of the cars generated for each direction of travel and which reach the intersection is 100 cars per minute in the ideal conditions proposed. The arrangement of the cars is equal in each lane and direction of travel so that there is no possibility of differentiated interpretation of the data obtained. The traffic lights communicate simultaneously, depending on the displays of the other traffic lighting systems and the emitted antennas.
Traffic lights are set to different time categories with a duration ranging from 20 s waiting time at the traffic light to 90 s in cases where traffic is low.
4.2. Simulation Structure of Traffic Flow
All these elements were simulated and analyzed in several stages in order to establish the ideal times and the density of vehicles that can travel in the area without the possibility of traffic congestion. Given the total transit capacity of the area and the way in which the road infrastructure was designed, it cannot withstand a vehicle density of more than 700 vehicles within 1 min. Thus, any delay between 3 and 5 s can lead to a blockage, which turns into a widespread congestion on the other main streets. Therefore, the scenarios were analyzed from the point of view of traffic lights established time, and in a secondary proposal, these are modified depending on the degree of congestion on the road segment. In simulated scenarios based on real data and problems facing this area, we cannot fully control congestion for both directions, the network is not unidirectional, and existing multiple bands on roads are a factor that makes it difficult to solve the problem only from road infrastructure, but we can reduce congestion and gridlocks.
The main goal is to present real solutions and show how important a traffic management system is based on hybrid communications like DSRC, V2X, V2V, and the newer VLC [
25,
26,
27], transforming and improving the prohibitions and behavior of vehicles in the vicinity of the area. Thus, some cars can avoid the area if they are not conditioned by a certain problem, using an alternative route, a temporary way to transit the congested road sector.
The problem is entirely conditioned by the shape and degree of congestion we have in the analyzed area, and these can be decision factors in the traffic control strategy. We can propose one-way traffic control strategies, in both directions, but also a tree-type control strategy. The latter type of control is proposed in a last simulation scheme represented in
Figure 3, showing that the route from the incident area to the shortest exit is largely similar to the branches of a tree [
28]. The complexity and variety of artificial control measures can lead to the structural peripheries of congestion by controlling the congestion tree. Imposing more than one road restriction through management systems and exposing them in real time for a period of time will lead to traffic jams and blockages, as most of those who transit the area and have embedded a certain procedure will have adaptation deficiencies. Behavior in clearing traffic by modern means of traffic management can help but can also affect the real situation [
29].
These decisions are adapted in time and the usefulness is constantly checked, such as the introduction of real-time road selection for each direction of travel and the traffic light system being guided by the flow of vehicles [
30].
Each simulated scenario presents a real situation and a simulative situation with a greater number of seconds for the time parked at the traffic light and the flow of vehicles that have transited the analyzed area, automatically outlining the congested areas. The flow of cars was generated starting from about 300 cars per minute given that the part of the road, according to the analyzed traffic information, allows a continuous flow of about 780 cars. In a final scenario, we have presented an extreme case that leads to a total blockage.
Figure 4 is the simulated case in which a shorter time for parking vehicles and an efficient traffic light system is recommended, so you can see the degree of congestion in
Figure 5, where the actual case where the main traffic light in the forward and left directions is approximately 86 s is exposed. The same scenario was simulated in
Figure 6, for a stationary time of 30 s at the main traffic light, the flow of vehicles generated a total passage of 605 vehicles without the formation of a queue or unforeseen blockages. In
Figure 7, the real case shows signs that in a long time with a constant flow of cars, queues can occur. The degree of congestion and how a blockage acts can be seen starting with
Figure 8, which shows signs of a traffic jam, although the number of cars that managed to leave the intersection is high without encountering problems in transit in the area. Thus, the degree of congestion in scenario 3 is the real version, in which the number of cars is about 500 per minute in all directions and all directions begin to congest the area and show a first blockage factor caused by a poor traffic and management system, given that the main boulevards have priority and that area is transited to connect the two main arteries (see
Figure 9). With the help of travel alternatives and bypasses combined with an alternative traffic management system, a fluidization up to 30% more efficient than in the absence of these solutions can be obtained, remedying the previously exposed problems, a percentage that would automatically avoid congestion of the road section. In scenario 4, both
Figure 10 as well as
Figure 11 have a degree of congestion exceeding the transit limits, with a blockage and queues for all directions. We can see from this scenario that an alternative route that could break the blockage is urgently needed and an intelligent traffic management system with information panels with the traffic situation in the area is a viable solution.
4.3. Simulated Scenario Results
According to the scenarios, we can simulate the observation of an average flow for their density, and for a high flow, there are several indicators that influence traffic conditions and lead to network congestion. For each of the four scenarios, the total transit capacity for the drum sector was made or analyzed, but also the definition of some parameters can be kept in consideration in the current conditions. Thus, we start with scenarios based on a flow of approximately 100 vehicles that pay from hubs related to each direction of travel, analyzing for each condition the total number of vehicles that cross the area without a total blockage. We can see that it is possible that the number of cars will generate an increase, and the rate of travel of the route in conditions for efficient traffic lights is satisfactory. At the same time, we need 90 s involved scenario care, which involves practical application and automatically generates their lock. Thus, the mentioned route can have an active flow of 700 cars in transit within 9 min in ideal conditions, otherwise, the area becomes congested for a calculated time between 15 and 25 min to the decongestion areas. This can cause the intersection of University Street with the two main boulevards (G. Enescu Boulevard and Ștefan Cel Mare Boulevard), with these being of priority, and it can have a much higher flow of vehicles exceeding 2000 cars in transit. For a better analysis and exemplification of similar settings, see
Table 1 below.
According to the data in the table, we can see that the current situation implemented in simulated scenarios reflects a degree of congestion that exceeds 80%, especially when the transit capacity of the analyzed sector is exceeded by the number of cars. Thus, the most important hours in traffic congestion are those mentioned earlier in the article, and for better coordination and streamlining of traffic, some experimental practical solutions will be proposed both through simulated elements and by improving the traffic management system and traffic lights.
The simulated scenarios aimed to test the real information and compare them according to certain factors that can negatively or positively influence the congestion and spread of the blockage both on the main street and on the adjacent ones or main boulevards.
4.4. Traffic Flow Solution and Three-Dimensional Projection
In order to solve most of the problems, we analyzed the triggers in case of congestion and tried to limit the spread of blockages in the form of spider webs, which would outline a traffic disability that can be maintained for several hours. The traffic flows are intense on the main arteries, more precisely the boulevards, and the connection between these two areas is made through the analyzed road sector, with this being limited as at this moment, there is only the possibility to change the direction in two areas leading to the higher education institution or to an agri-food complex. The main problem is that the part of the road does not benefit from an alternative route to bypass the congested area, so in the following simulation, the diagram of the area was graphically reproduced with the remodeling of a road passage that allows turning right in two areas so that the return to the boulevard is possible, thus isolating the blockage. In the same area, at the level of simulation and at the level of local infrastructure, three info-traffic panels are located that highlight the degree of congestion in real time with the help of surveillance cameras already installed at the intersection. This information is ideal for the categories of drivers who want to leave the boulevard and make the right turn towards University Street. These are real and implementable solutions based on the current infrastructure without influencing the driving conditions and without creating problems for the inhabitants by adapting to a new technology [
31].
For a better coordination of things within the application, vehicles were prioritized, transforming the car network into a series of autonomous cars, connecting each line, synchronizing everything with the road lighting system so the congestion problem is solved in large proportions in cases in which vehicles are equipped with systems dedicated to communication with DSRC, V2X, V2V, or VLC infrastructure [
32,
33], either from the factory or individually developed equipment.
In this sense, our research team is working on the development of prototypes of communication through visible light and radio frequency, devices capable of capturing and processing information from the external environment and then transmitting it to other vehicles for communication and formation of an autonomous network dedicated to traffic safety.
Figure 12 represents the design of a model after the respective area with changes in the central part of the traffic area, the delimitation of directions, and the formation of an alveolus dedicated to returning to a main road sector, which fluidizes traffic and compresses congestion only at traffic lights, which are of a very short time. Thus, the data packet function was designed to provide approximately 500 vehicles for all directions, synchronizing with the road infrastructure and the priorities imposed depending on the traffic situation.
The simulation process was allowed to run within 1–3 h, but also for a period of 12 h to provide a broader perspective on the average parking time, the number of cars that crossed the area, the occurrence of blockages, and the presence of traffic events. The technical solutions that refer to the management of traffic management systems are effective, especially if in this intersection the changes presented in
Figure 13 will be made, where the alveolus delimiting the directions of travel was highlighted, making it possible to reclassify traffic on a secondary artery without pushing cars into a queue that would turn into a branched congestion.
Thus, the proposed change increases the flow capacity for priority traffic and offers the possibility to avoid the congested area, about 70% of the current traffic flow could benefit from this solution and would automatically increase the transit capacity through that area. Between the two boulevards, the road section has about 900 m and the capacity of 700–800 cars that can transit the area within 10 min, provided that the traffic systems are set below 30 s and the secondary streets do not impose limitations and turns. The current proposal comes with a plus in order to increase the capacity of vehicles transiting the area. According to the graphs shown in
Figure 14, there is no downtime in parking vehicles, and there is constant speed, a low waiting time, and a lack of congestion.
The designed logic diagram is highlighted in
Figure 15 and provides an overview of the entire simulated process with a perspective on the total number of vehicles left at each point and the total number of vehicles that passed through the designed nodes.
According to the results obtained, the congestion location is a problem generated by the lack of right-of-way priorities to avoid the area, but also a poor road management that starts from the main boulevards with a higher flow of cars. In a first phase, the proposed solution is a topical one based on the current infrastructure [
34]. The future model for implementation also includes a practical part of traffic based on the use of devices for traffic management and communication technologies such as V2X, V2I, V2V, and DSRC [
35,
36]. Installing a road-side unit (RSU) and several onboard units (OBU) on cars and monitoring traffic through modern means of control will be able to demonstrate the usefulness and need for an intelligent traffic management system, and also the need for autonomous cars [
37,
38]. Some information may differ in a real case because the total unpredictable and anticipated human factor intervenes, and any simulated model in this case is exposed and the rule becomes invalid, so we consider the presence of dedicated systems useful, installed in both heavily trafficked areas as well as in cars.
In ideal cases, a congestion can be remedied in a relatively short time of about 7–10 min, but these times are in accordance with simulations that do not involve other events. In total unforeseen traffic, they can negatively affect the situation and congestion, having a duration exceeding 30–50 min to be resolved. That is why an alternative route dedicated strictly to the areas with urgent problems is necessary because such a shortcut can save time, lives, and can reduce waiting times and dispel traffic jams [
39].
5. Conclusions
Accidents and loss of life caused by poor traffic management and poor adaptation of speed to weather or traffic conditions cause both accidents and blockages. In this paper, the traffic congestion caused by various reasons was analyzed and presented, simulating several models, dividing the traffic according to certain criteria and prioritizing certain areas. These elements were made based on a real case and based on the information received on the density of cars in that area. Following the analysis and simulations, we observed the decision factor in creating congestion and how it develops and branches, making the area impassable. Thus, the analyzed and subsequently simulated elements present the way in which strategies can be proposed to streamline the process and become feasible in the near future. According to the results obtained and the evolution of the results from the paper, we can see that the car network and the blocked route have been improved by technical implementations or by elements of modern road infrastructure. Suceava, the city of our study, is not a choice in itself, but rather a choice given by the symmetry of our trajectories and the fact that we see the problems and want to act on them. The study can be scaled up to other cities as well to the detriment of freight transport by taking into account road infrastructure. As a consequence, we analyzed real cases and scenarios, for which we found solutions with the help of simulations for an efficient traffic management regardless of the nature of the events. The article presents enough elements that constitute a starting point for the application of simulated models and methods in practical cases, but also in the advancement of intelligent transport systems for congestion control, intersection control, signaling and display of traffic flow at busy intersections, and orientation driver information [
40].
We can also advance and develop more generalized methods of traffic prediction and temporal and temporary analyses of vehicle flow through systems specially created in this direction using elements and communication methods dedicated to safe driving. The next step in the development of this direction is the simulation in a real framework of a network of vehicles based on V2X communications and a DSRC system, using a complete system consisting of two or more OBUs and an RSU for control, command, and analysis of data. We want to attach to this simulation the VLC system developed within a research project dedicated to the automotive sector and which could create a mix of technologies capable of providing a complete solution. We consider this approach necessary and we want to prove the usefulness and necessity of dedicated systems for transport and road safety by putting them into practice in real situations. Such systems and methods used in the simulations presented in this article can be applied anywhere in a problem related to traffic congestion. Whether we are talking about traffic density congestion or road accidents and events, the use of alternative routes or intelligent transport systems with information panels at each main artery would streamline traffic without major events. Communications and applications dedicated to the automotive sector have huge potential and benefit from the current state of technology and existing infrastructure, so the implementation costs of the various systems become much easier to achieve and at much lower costs. The complementarity of systems based on VLC-V2X technologies adds the advantage of non-interference with the radio communications of any kind, excluding interference with other mobile communications and DSRC, being very useful in urban areas where there is a saturation in radio frequency.
The use of intelligent transport systems is discussed a little in this article, as the development of a solution to help this problem is underway and provides an efficient combination of the methods addressed in this article by creating dynamic vehicular networks. All the authors’ efforts are channeled in the mentioned direction and support the efficiency in terms of traffic management, intelligent traffic systems, and autonomous equipment dedicated to the automotive sector.