The behavior of drivers and the applied driving operations affect fuel consumption and gas emissions for each trip. Several protocols have been proposed in the literature aiming to assist drivers and recommend the best operations to enhance the efficiency of driving trips. These protocols can be easily categorized based on the targeted road network: downtown and highways. Moreover, the driving behavior is highly affected by the context and requirements of each scenario. Figure 1 graphically illustrates the main considerations of efficient traffic control protocols in downtown scenarios.

First, as seen from the figure, finding the efficient path between a source and a destination on a grid-layout downtown road network depends on several real-time traffic characteristics. These include the locations of the source and destination, the estimated travel time of the path chosen, required fuel consumption and gas emissions, and the distribution of traffic and congestion. Second, the located traffic lights must be operating efficiently. This is done by considering the traffic density of the competing traffic flows, their context, and required fuel and gas emissions.

In contrast, Figure 2 illustrates the primary considerations of efficient traffic control protocols on highways. As we can see from the figure, the main considerations of these protocols are the shape of the highway design, the speed of vehicles, traffic congestion, and driving behaviors, including lane changes and exit/entrance points.

First, the grid layout of downtown areas provides several route options that drivers can follow towards their targeted destinations. Selecting the most efficient route that considers the real-time traffic characteristics of the investigated area of interest has been intensively investigated by several studies and projects. Finding the fastest path that leads towards the targeted destination has been developed as essential phase in this field [25,26].

The fuel consumption and gas emission parameters have also been investigated for the selected and candidate routes to recommend the most efficient selection. Mohammad et al. [27] measured the estimated required fuel consumption for the shortest k-routes towards the targeted destination. The estimation process considered the traveling distance and speed of vehicles in addition to other real-time traffic characteristics using the IFC model [20]. The route that was estimated to consume the least amount of fuel is recommended as most efficient route. Xu et al. [28] have applied the same idea, however, they have estimated the fuel consumption for the k-fastest routes instead of the closest one.

Bani Younes and Boukerche [25] introduced a path recommendation protocol that measures the traveling distance and traveling time of each road segment on the grid layout of a downtown road network. A balanced route that considers the traveling time and traveling distance between the source and destination is then selected. No drastic delay or traveled distance is experienced in the selected balanced route towards the targeted destination, and it is constructed based on the characteristics of the linked road segments. Measuring the fuel consumption and gas emission of the balanced route, the fastest route and the shortest route using an instantaneous fuel consumption model is selected [20]. Comparing the fuel and gas emissions of vehicles following these routes, the obtained balanced route can be recommended as the efficient route, as traveling vehicles on this route require the least amount of fuel.

Furthermore, many eco-path protocols have been designed to efficiently control the amount of consumed fuel and produced gases during each vehicle’s trip. Kono et al. [29] proposed an ecological path recommendation protocol. This protocol used Dijkstra’s algorithm to find the path that requires the least amount of fuel. It depends on a centralized gathering of data for the required fuel to traverse each road segment on the investigated downtown network. Chang et al. [30] developed VANET-based A* route planning algorithm to find the fastest and most efficient route in terms of fuel consumption. This is based on two main real-time traffic sources: traffic data of road segments that the vehicle passes through and traffic information provided by Google Maps.

TraffCon [31], eCo-Move [32], and EcoTrec [33] are smart solutions aimed at enhancing the efficiency of traveling vehicles. TraffCon is an efficient algorithm for vehicle routing that aims to reduce the traveling time of vehicle trips and decrease fuel and gas parameters [31]. It considers three main real-time parameters to recommend a certain route: traveling time, used capacity of the road network, and fuel consumption. The eCo-Move system [32] is constructed based on the assumption that “there is a theoretical minimum energy consumption. That is achieved with the perfect eco-driver travelling through the perfectly eco-managed road network”. Thus, a microscopic simulation environment is implemented with installed eCo-Move applications as realistically as possible. This aimed to test and recommend the most efficient behavior and most efficient road management conditions to reduce the fuel consumption of each selected route. EcoTrec [33] is an eco-friendly algorithm that finds the most efficient route considering the fuel consumption parameter. It mainly utilizes the efficiency of selecting individual road segments and considers a number of factors. It balances relevant factors of traffic conditions over the investigated road network such as travel time, road congestion level, and gas emissions. This enhances the fuel consumption and the efficiency for the selected route.

Table 1 summarizes the main considerations and characteristics of some ecological path recommendation protocols and algorithms that have been proposed for efficient traveling over downtown areas. The input and output of each protocol are illustrated in the table.

Existing road intersections in a downtown road architecture are shared among several conflicting traffic flows. These intersections are usually controlled by stop signs, roundabouts, or traffic lights aiming to safely schedule competing traffic flows. Few driving assistant protocols have investigated the traffic characteristics and the efficient options that drivers can take around road intersections that are controlled by stop signs and roundabouts [34,35]. However, the fuel consumption and gas emission parameters have not been directly investigated or studied at these intersections.

Traffic lights have been considered as the most sophisticated solution to control conflicting traffic flows at road intersections. Several techniques and technologies have been utilized that aim to provide efficient schedules there [36,37]. These solutions aimed to increase the throughput of the investigated road intersections and decrease the waiting delay time of traveling vehicles. Moreover, many research studies have intelligently considered the parameters of fuel consumption and gas emissions when setting the schedule of each located traffic light [38,39,40]. A recent infrastructure-less traffic control system that is solely based on vehicle-to-vehicle (V2V) communicationswas proposed by Ferriro et al. [41] for downtown road networks.

Virtual traffic lights are introduced in this system, which aims to control the competing traffic flows at road intersections while considering their real-time traffic characteristics. Ferriro and d’Orey [42] proved the impact of these traffic lights on carbon (CO2) emissions mitigation. However, Vlasov et al. [43] proposed an adaptive traffic light control algorithm that mainly targets reducing the fuel consumption and gas emissions of traveling vehicles. It considers the dynamic transport demand on the road intersection besides real time traffic characteristics of the competing traffic flows.

Assisting drivers to take the most efficient action and prepare for upcoming conditions help reduce fuel consumption as well. Haritenstien et al. [39] investigated the effects of gear choice and the distance between each vehicle and the traffic light when drivers received the stopping or passing signals on the efficiency of that vehicle in terms of fuel consumption. This assessment has been applied for a large-scale simulation to obtain more accurate and beneficial results. Moreover, Ngo et al. [44] introduced an adaptive traffic light scheduling algorithm and optimal speed advisory method. It recommends the most efficient speed to each driver in order to reduce the total fuel consumption and gas emissions of traffic at the investigated area of interest. The scheduling algorithm can be adapted to solve the yellow-light-dilemma problem. This would extend the yellow signal time for vehicles in a zone, which gives them time to safely stop or pass through the intersection. This increases the smoothness of movement of the vehicle and enhances the efficiency parameters.

Intelligent algorithms have been utilized to set the optimal schedule of traffic lights. These scheduling algorithms also aim to reduce the emission and fuel consumption of the traffic at the investigated area. Alba [40] used particle swarm optimization techniques. Furthermore, Soon et al. [45] developed a pheromone-based green transportation system with three comprehensive phases to tackle all high fuel consumption scenarios in the downtown area. These include: traffic congestion prediction, coordinated traffic light control strategy, and cooperative green vehicle routing. The integration between the traffic light scheduling and the routing phases helps to find the route with the least fuel consumption.

Furthermore, several researchers have considered the context of the traffic while selecting the efficient schedule of the located traffic lights. Younes and Boukerche [46] have considered the existence of emergency vehicles and public transportation. They assigned a higher priority for these vehicles to pass through the signalized intersection quickly and safely. Salin [47] considered public transportation vehicles and assigned them a higher priority to pass through the intersection. Investigating the efficient parameters of these algorithms have shown reduction in the fuel consumption and gas emission parameters. However, Suthaputchakun and Sun [38] considered heavy loaded vehicles with a higher priority to pass through the signalized road intersections. These vehicles consume more fuel compare to normal vehicles due to braking, stopping, and restarting actions. The latter algorithm proved an enhancement in the efficiency of traffic in the investigated area of interest. Table 2 summarizes the details of the main techniques used to efficiently control the traffic light schedule. The input and output of each algorithm have been listed in the table.

The straight, wide, and multiple-lane design of highways is considered an efficient road design. Vehicles consume less fuel to travel on highways compared to the rate of fuel consumption in downtown scenarios. The most common traffic distributions on highways are platooning (i.e., set of platoons where vehicles in each platoon are traveling closely at a steady speed). Each platoon contains several vehicles that follow each other with short gaps and times [48]. However, the comfortable flat design of highways encourages drivers to increase their speed and apply a high acceleration rate. They may also need to apply hard braking in some cases to avoid obstacles or take an exit. These behaviors drastically increase the rate of fuel consumption and gas emissions on highways.

Theoretically, several studies have documented efficient driving tips on highways such as smooth acceleration/deceleration, optimal gear shifting, anticipating traffic, avoid idling, etc. [7,49]. Many environmental organizations have claimed high reduction in the fuel consumption and gas emissions by following some eco-driving techniques compared to the average driving style [13]. Some countries and environmental organizations have introduced eco-driving courses. These courses aim to educate and train drivers for efficient driving [13]. The impact of eco-driving training courses on fuel consumption have been investigated by measuring the historical fuel consumption and gas emissions rates of vehicles during a certain period of training time [50]. The reduction rate in these studies depends mainly on the individual skills of the involved drivers and the design of the tested area.

Furthermore, traffic congestion, accidents, and obstacles cause driving behavior to change on highways. Providing drivers with real-time efficient recommendations such as speed, acceleration rate, or the optimal gear selection should enhance efficient driving behaviors. He and Wu [51] introduced eco-driving advisory strategies that enhance the efficiency of driving on highways, mainly by recommending the optimal speed for each platoon of vehicles. Lee and Son [52] recommend the most efficient depth of the acceleration/deceleration pedal according to the selected gear option. In addition, they correlated the angle of the steering wheel to the fuel consumption rate on highways as an efficient measure of driving.

Additionally, advanced design technologies in vehicles have directly enhanced the efficiency conditions on highways in terms of reducing fuel consumption and gas emissions [53]. For instantce, cruise control equipment has been added to most recent vehicles. Steady speed can be easily maintained on highways using this equipment [54]. Connected vehicles that communicate through vehicular ad-hoc networks have contributed as well to enhancing traffic efficiency. Ploeg et al. [55], Taiebat et al. [56], and Wang et al. [57] have developed different eco-cooperative adaptive cruise control protocols. These protocols have used the connecting technology among traveling vehicles to obtain the most efficient traveling speed according to the real-time traffic distribution and road design. High rates of reduction in the fuel consumption and gas emissions have been reported for utilizing these protocols.

In addition, autonomous and electronic vehicles are foreseen as future development technologies. Several research studies and industrial organizations are working towards them [53]. Autonomous vehicles have several promising features in terms of enhancing the safety conditions on road networks. Moreover, they should enhance the efficiency because the efficient driving conditions are applied automatically without human interference. Several electronic efficiency control systems have been developed to enhance the performance of autonomous and electronic vehicles [55,56,58]. Highways have been the most suitable environment to verify these protocols [59]. Table 3 illustrates the details of main highway eco-driving techniques including the required input for each protocol and its obtained results.