AI and Non-AI implementation of eco-driving techniques. [IoT]


Transport is a vital part of our modern economies as the quality of most of our life depends on the modern, accessible and efficient transport system. The development of the transport industry comes with costly environmental repercussions. Indeed, it contributes heavily to air pollution, noise but most importantly to climate change. According to the European Environment Agency, the transport industry is the biggest source of carbon emission. It contributes to 27% of the EU’s total carbon footprint of with 70% are issued from cars [1].  Ergo, the European Union put a plan focus on decarbonizing transport with a goal set towards ‘net-zero’ greenhouse gas across the euro Zone by 2050 [1].  It entails the reduction of greenhouse gas emissions by 24% in 2020 and by 32% by 2030 [2]. One solution implemented to reach that goal is the introduction and promotion of eco-driving.


What is Eco-driving?


Eco-driving is an attempt to change people’s driving behavior through so-called advice such as shifting gears sooner, driving at high-gear with low speed, smooth driving by anticipating traffic, avoiding sharp acceleration/deceleration, among others [3]. Those rules can theoretically lead to the reduction of greenhouse gas emissions, the diminution of fuel consumption, and a decrease in the number of road accidents [4].


The application of eco-driving techniques is divided into multiple ways. In the next section, we will explore the application of eco-driving at the operation level, on the choices of vehicles, and the choice of route.


How is Eco-driving applied at the operation level?


The operation level relates to the behavior of the driver during a drive [5]. The advancements at the operation level aimed at reminding the drivers to drive sustainability. Ergo, a lot of focus was on the implementation of devices that would provide real-time or trip-summarized eco-driving advice during or after trips. It would allow drivers to adjust their driving behavior for the maximization of fuel economy.


What are some AI-based applications that help with eco-driving?


An example of such a project is from the project GamECAR. This EU-funded project aimed at motivating the promotion of eco-driving through a multiplayer gaming platform allowing the drivers to set missions and invite others to participate collaboratively or competitively[6][7]. Using visualization techniques such as Augmented reality, they managed to monitor the eco-driving score evolution while providing the drivers with a personalized plan for improvement. [8][9]


Example of Ai-based eco-driving apps


Other AI-based applications related to eco-driving were developed, and you can see the illustration of a few on them in figure \ref{fig: subfigures}. For example, we have a few android applications that could display second-to-seconds eco-score ranging from ‘bad’ to ‘very good. Those android applications are composed of three fuzzified variables. They are velocity, road slope, and power consumption [10]. Going a step further, a smartphone application called DrivingCoach can detect the users driving behavior and pattern and then suggest new behaviors to reduce your fuel consumption in real-time [11].  In this application, 8 variables have been used in a fuzzy way to provide feedback on the user’s driving.


Another implementation of operation-level apparatuses is a device that allowed the introduction of eco-driving feedback that provided real-time trip emission information [12]. It used a color scheme to show real-time feedback of both the fuel consumption and CO2 emission. The information is displayed on a dashboard, showing features like the average speed, the travel time, the fuel consumption levels along with C02 emission. Using those devices on a small sample of drivers within a city showed an average improvement of 6% regarding fuel economy. [5][13]


How does eco-driving relate to vehicle choice?


The vehicle choice relates to the type of vehicle you are driving and the emission levels produced by the energy released by its motor and the equipment within the vehicle. Before buying a car, knowing if a said-car is eco-friendly, plays a major role in the driver’s potential consumption levels. A lot of policies/regulations have been implemented by the EU, over the past two decades, to provide a more eco-friendly automobile market.


What are eco-cars?

The definition of eco-cars is:

  • Vehicles with alternative fuels such as ethanol with energy consumption:
    • 9.7 m3 CNG/100 km
    •  9.2 l/100km
  • 37 kWh electric energy/100 km
  •  Vehicles with conventional fuels (hybrids included) with CO2 emission less than:
    • 120 g/km
    • PM < 5 mg/km for diesel-powered vehicles


A lot of governmental policies have pushed consumers towards buying a more eco-friendly car. For example, in the EU, if the average CO2 emission of a car exceeds the maximum C02 threshold in a year, the manufacturer has to pay a fee called excess emission premium for every car registered in the market [15]. That fee varies from :

  •  €5 for the first g/km of exceeding
  • €15 for the second g/km
  •  €25 for the third g/km
  • €95 for each subsequent g/km.


What are some advantages of buying an eco-car?

In Japan, the central government imposed tax incentives called the Green Tax Scheme. These taxes give incentives for buying low emission and fuel-efficient cars, which are identified as such by a certification process [14]. In Germany, measures to promote electric mobility were put in place. Such advantages include a tax break for zero-emission vehicles and policy measures such as special lanes or parking spots for zero-emitting cars. In Sweden, an incentive scheme that depends on several performance parameters has been applied. As a result, a person may get up to ~1100€ for the registration of an eco-car meeting environmental criteria [14].


Besides the benefits various governments started to implement when buying a new car, the choice of vehicle has other factors to take into consideration. Things like vehicle weight, tire pressure, aerodynamic drag, and vehicle maintenance should affect the driver’s decision to buy a particular car. Indeed, the weight of the vehicle should be minimized. The lighter the car is, the less energy is required to move, which translates into lower fuel consumption. The same principle applies to the general maintenance of the car and its components. On top of everything, minimizing aerodynamic drag, like fully closing the windows or avoiding putting external cargo, will decrease the air resistance on the car ergo, decreasing the amount of energy required during the drive [16].


How can Route Choice be improved by eco-driving?

Route planning relates to planning what road to take before the drive taking into account things like traffic congestion, time of the day, etc. [17]. Mapping applications, such as Fueoogle, were developed to address the route choice. It uses a participatory sensing service. This service maps vehicular fuel consumption to enable drivers to figure out the most efficient route for their vehicles between two endpoints [18].


With more recent implications, the term eco-route was created to recommend the driver’s cost-effective routes in real-time. The idea behind the term is that one can decrease time-traveled to maximize lower fuel consumption [19]. The obtention of those eco-routes is calculated through various AI techniques. It includes graph theory principles where the nodes represent junctions, the edges are roads, and the costs estimated by the energy/fuel needed to travel between two connected nodes [19].


Eco-routing navigation systems have now been developed to fulfill that purposed [20]. They are composed of a connector to a Dynamic Roadway Network database. It is a roadway network integrating real-time and historical information about traffic from various data sources with an embedded data fusion algorithm. The DRN is paired with an emission parameter set, which is a compilation of emission factors. Those factors vary depending on the types of vehicles, road characteristics, and conditions of traffic.


The final component of the routing navigation system is the routing engine (RE). The RE is the optimal shortest path algorithm used to calculate the best eco-route. Finally, The optimal eco-route is displayed through a user interface.





How effective are eco-driving training programs?

The goal of an eco-driving training program is to train experienced or new drivers to become sustainable drivers. Indeed, strategies and studies have been put in place to test the efficiency of the training programs. The figurere below summarizes a series of published eco-driving training programs and their effect on fuel consumption before or after straining. Generally, we can see that those training programs help reduce fuel consumption by 2-15%, which depends on the type of program [16]. This variation is mainly because each training program is different and varies in terms of strategies, driving conditions, country, etc. [21]


How are the eco-driving training programs? 

The training programs are usually divided into 3 steps. They are theoretical training, practical training, and a combination of both. For instance, a study shows the effect of simple advice and eco-driving training and how it changes driving behavior. As a result, it concluded that the fuel consumption decrease by about 12% when given simple eco-driving advice, which was higher than the full eco-driving training which experiences a decrease in fuel consumption by 11% [22].


Another study compared the effectiveness of online learning and hard-copies. Additionally, they tested giving eco-advice on 2 hours of driving lessons on online learning platforms, 1.5 hours of driving lessons and, half a day of driving workshops [23]. The result showed that all the above actions had a positive effect on fuel economy. But, there was no significant difference between them [23].


Another paper compared the results of two eco-driving strategies: an in-car feedback system and a feedback system with a personal trainer. The results concluded that both resulted in a fuel saving of about 6.8%, but no difference was observed between the two strategies [24]. Controversially, another study reported that solely theoretical training did not have any effect on either the short-term or the long-term driving behavior, it stated that practical training is necessary [25].


efficiency of training programs on fuel reductions


What is the effect of eco-driving training programs on driver’s behavior?

In general, the results obtained by most studies are usually recorded straight after a set experiment which usually shows a positive outlook. However, long-term studies show that the eco-driving skills tend to disappear over time. It is, mainly, because it is generally hard for someone to change their driving habits, especially if one has been driving a particular way for years [16] [26] [27].


Additionally, other factors influence fuel consumption meaning that the training result may be affected by it. I will go through those factors later in the paper. A survey showed that eco-driving practical learning was more effective than a high level of theoretical learning.[28]. Additionally, eco-driving training was more efficient in cities rather than on highways. Similarly, training received better results for manual transmission instead of automatic transmission cars [27].


On average, most studies used existing experienced drivers, with very few considering new learning. The ECOWILL European project was such a study that focused its effort on new drivers for 3 years in 13 different European countries [29]. That training program provided eco-driving seminars for both experienced and newcomers. The new drivers were taught standard driving techniques that include the golden and silver rules of eco-driving.


The ECOWILL project aimed at educating 10 million learners, and beginner drivers with the principles of eco-driving. As well, the long-lasting benefits sustainable driving has. By the end of the project, the European Commission made eco-driving a mandatory component of the driver’s test in all states in the European Union [29]. As a result, studies have shown that eco-driving for learners has been instated and shaped through their education. However, the understanding of the experienced drivers was broader, which included strategic to tactical decisions, made adjusting to eco-driving techniques harder [27].


A Last Word

To conclude, there was a small overview on how eco-driving was implemented at operation, route choice, vehicle choice, and training programs. It has proven that eco-driving allows you to save on your fuel consumption. A lot of AI/IoT and non-AI-based techniques attempt to promote this concept of eco-driving.

If you made this far in the article, thank you very much.


I hope this information was of use to you. 


Feel free to use any information from this page. I’d appreciate it if you can simply link to this article as the source. If you have any additional questions, you can reach out to If you want more content like this, join my email list to receive the latest articles. I promise I do not spam. 





If you liked this article, maybe you will like these too.

Google Analytics: How to get your Top N Products by location?

What is the difference between Data Scientist, Data Analyst, Big Data and Machine Learning Specialist?

Unfolding the Mystery of the Zipf’s Law



[1] Agency, E. E. Laying the foundations for greener transport (2019).
[2] Commission, E. A policy framework for climate and energy in the period from 2020 to 2030 (2014).
[3] Andry Rakotonirain, G. S.-P. P. D., Narelle Haworth. Research issues in eco-driving (2011).
[4] Sinisa Husnjak, T. B., Ivan Forenbacher. Evaluation of eco-driving using smart mobile devices (2015).
[5] chen chen, Y. Y. Y. Z. J. R., Xiaohua Zhao and Liu, X. Driver’s eco-driving behavior evaluation modeling based on driving events (2018).
[6] Aris S. Lalos, K. M., Stavros Nousias. Gamecar: Gamifying self-management of eco-driving. ercim news (2018).
[7] Vlachos, E., Lalos, A., Berberidis, K., and Tselios, C. Autonomous driving in 5g: Mitigating interference in of dm-based vehicular communications. pp. 1–6 (2017). doi:10.1109/CAMAD.2017.8031619.
[8] University of Patras, B. M. S. F. p. p. d. l. i. i. y. d. t. e. l. i. d. a. d. G. S. W. L. F. I. o. S., University of Leeds, Technology for Transport, D., and Networks, K. S. s. Gameecar (2018).
[9] Nousias, S., et al. Managing nonuniformities and uncertainties in vehicleoriented sensor data over next generation networks (2018). doi:10.1109/PERCOMW.2018.8480342.
[10] Frank, R., Castignani, G., Schmitz, R., and Engel, T. A novel ecodriving application to reduce energy consumption of electric vehicles. In 2013 International Conference on Connected Vehicles and Expo (ICCVE), pp. 283–288 (2013). doi:10.1109/ICCVE.2013.6799807.
[11] Araújo, R., Igreja, , de Castro, R., and Araújo, R. E. Driving coach:
A smartphone application to evaluate driving efficient patterns. In 2012 IEEE Intelligent Vehicles Symposium, pp. 1005–1010 (2012). doi:10.1109/IVS.2012.6232304.
[12] Boriboonsomsin, V. A. . B. M., K. Eco-driving: Pilot evaluation of driving behavior changes among u.s. drivers. (2010).
[13] Danijela Baric, M. P., Goran Zovak. Effects of eco-drive education on the reduction of fuel consumption and co2 emissions (2013).
[14] UNECE. Environmentally friendly vehicles and the world forum for the
harmonization of vehicle regulations. (2012). 
[15] Reducing co2 emissions from passenger cars. Accessed: 2021-03-22. 
[16] Huang, Y. Eco-driving technology for logistics transport fleet to reduce fuel consumption and emissions. (2018), 27.
[17] chen chen, Y. Y. Y. Z. J. R., Xiaohua Zhao and Liu, X. Driver’s eco-driving behavior evaluation modeling based on driving events (2018).
[18] Pham, N., Ganti, R., Nangia, S., Pongthawornkamol, T., Abdelzaher, T., Heo, J., Khan, M., and Ahmadi, H. Fueoogle: A participatory sensing fuel-efficient maps application. (2009).
[19] Kubicka, M., Klusacek, J., Sciarretta, A., Arben, C., Mounier, H., Thibault, L., and Niculescu, S. Performance of current eco-routing methods. pp. 472–477 (2016). doi:10.1109/IVS.2016.7535428.
[20] Boriboonsomsin, K., Barth, M. J., Zhu, W., and Vu, A. Eco-routing navigation system based on multisource historical and real-time traffic information. IEEE Transactions on Intelligent Transportation Systems, 13 (2012), 1694. doi:10.1109/TITS.2012.2204051.
[21] Gardelis, K., Lalos, A., and Moustakas, K. Development of an ecodriving simulation training system with natural and haptic interaction in virtual reality environments (2018). doi:10.5220/0006619500940101.
[22] Cindie Andrieu, G. S. Comparing effects of eco-driving training and simple advices on driving behavior (2012).
[23] Jeffreys, I., Graves, G., and Roth, M. Evaluation of eco-driving training for vehicle fuel use and emission reduction: A case study in Australia. Transportation Research Part D: Transport and Environment, (2016). doi:10.1016/j.trd.2015.12.017.
[24] Strömberg, H. and Karlsson, M. Comparative effects of eco-driving initiatives aimed at urban bus drivers – results from a field trial. Transportation Research Part D: Transport and Environment, 22 (2013), 28–33. doi:10.1016/j.trd.2013.02.011.
[25] Schall, D., Wolf, M., and Mohnen, A. Do effects of theoretical training and rewards for energy-efficient behavior persist over time and interact? a natural field experiment on eco-driving in a company fleet:. Energy Policy, 97 (2016), 291. doi:10.1016/j.enpol.2016.07.008.
[26] af Wåhlberg, A. Short-term effects of training in economical driving: Passenger comfort and driver acceleration behavior. International Journal of Industrial Ergonomics – INT J IND ERGONOMIC, 36 (2006), 151. doi:10.1016/j.ergon.2005.10.001.
[27]  Barla, P., Gilbert-Gonthier, M., López Castro, M., and MirandaMoreno, L. Eco-driving training and fuel consumption: Impact, heterogeneity and sustainability. Energy Economics, 62 (2017). doi:10.1016/j.eneco.2016.12.018.
[28] Mansfield, L., Guros, F., Truxillo, D., and MacArthur, J. Individual and contextual variables enhance transfer for a workplace eco-driving intervention. Transportation Research Part F: Traffic Psychology and Behaviour, 37 (2016), 138. doi:10.1016/j.trf.2015.12.001.
[29] ECOWILL. Ecodriving short-duration training for licensed drivers and integration into driving education for learner drivers; experiences and results from the ecowill project (2013).







Leave a Comment