Human Mobility, Artificial Intelligence and Climate Change

This chapter presents in a concise and documented way the story of the climate change phenomenon and its anthropogenic origin. It presents the arguments put forward by those who deny that climate change is caused by human activities as well as the arguments of those who accept this notion and support the need for immediate action. Both arguments are presented with a summary of supporting data and documentation. Key data of CO₂ emissions per capita and per GDP are presented for some key polluting countries for a period of 30 years while data of actual quantities of CO₂ emitted in Gt/year are given for the six top polluting economies (China, USA, EU, Russia, India, Japan). The trends in the quantities of CO₂ emitted each year by the six most polluting countries, show a marked increase in the emissions after the year 2000 for some of these countries. The concise and accurate presentation and comments on emissions data and the reference to key reports and documents by various relevant organisations, is complimented by presentation of some of the most pronounced impacts of climate change that are already here. These are the major natural disasters that occurred around the world over the last decade which bring the most convincing evidence of the dire consequences that can be caused by climate change and global warming. The concluding sections of this chapter defend the notion that climate change is here, it is getting accelerated by human activities and their related GHG emissions, and that “doing nothing” is not an option. Already, the negative socioeconomic impacts of extreme weather events are felt in an increasing number of countries, and this should provide people and governments with enough evidence as to the painful consequences of a “do nothing” situation. It is argued that the consequences and impacts of “doing nothing” will be much bigger and more costly to humanity than the cost of the actions necessary to mitigate climate change and achieving full decarbonisation in the foreseeable future. The fight to combat climate change is seen as the only path in front of humanity to save the planet from the catastrophic consequences of a full-blown climate crisis.

In terms of its CO₂ emissions, the transport and mobility sector is one of the top (and in some countries, the top) polluting sector globally. It accounts for almost 25% of all CO₂ emissions globally and higher than that in certain areas such as around 28% for the EU, 27% for China, 28% for the US etc. This chapter looks exclusively at the relation of the transport sector with climate change and examines the issues involved and the key elements of interaction of the sector vis-à-vis the unfolding climate change crisis. Primarily, it aims to consider the transport sector as an emitter of Green House Gasses—GHG, and to assess the main policies for its decarbonisation. It also considers the incoming innovations in the sector and their potential climate friendliness. The emphasis is on road transportation because this is the most polluting mode by far. The other transport modes are also considered but, in less detail, and only when necessary. The highlight of this chapter is that it presents a concise synthesis of the various proposals and suggestions for decarbonisation measures and policies in various studies by national or international agencies and highlights the types of transport decarbonisation measures and policies that are most promising and likely to be widely accepted. These analytical presentations follow the triptych: “avoiding”, “shifting” and “improving/deploying” measures i.e., policies that aim to make the users of the transport system avoid making trips, shift to less emitting modes, and improve/deploy new mobility technologies that will enable low or no greenhouse gas emissions. Besides this analytical presentation of the necessary actions for transport decarbonisation, this chapter examines also the so called five pillars of future climate-friendly mobility, i.e. Electro-mobility (electric vehicle propulsion and battery storage systems), Automated mobility (driverless, intelligent vehicles communicating data and information with the infrastructure and other vehicles), Big data and mass information collecting and processing technologies, Artificial Intelligence (AI) and its many applications for mobility and transportation, and New cultural and socioeconomic mentalities that are developing in connection to the changing travel and mobility characteristics. The reader is then led through a first assessment of the “odds for success” i.e., an assessment of the measures and policies that are considered as enough to effect zero carbon emissions and their chances of been implemented in time to avoid global warming above the 2 °C by 2050 or’60. This is done by examining progress in implementing the various decarbonisation measures in the four categories: Managing the “demand” for travel; Improving the energy efficiency of conventional vehicles; Shifting passenger travel and freight to less carbon intensive modes; and, Transitioning to “clean” no-carbon emitting energy vectors. The chapter ends by stressing that all the different policies and measures for decarbonisation in the transport sector, must be implemented in a concerted and coordinated way as there is no “silver bullet” for the successful decarbonisation of the sector. In the uncertainty that reigns globally, the individual decarbonisation measures and initiatives that have been taken so far by governments around the world, are not considered to be enough.

This chapter begins with a review of existing research work on transport resilience in general and against climate change in particular. Understanding how transportation systems respond to and recover from the various disruptions, including the climate related ones, is crucial for developing effective strategies to enhance system resilience. This review reveals the various approaches and methodologies that researchers have employed to evaluate transport system vulnerability and adaptive capacity including in the face of climate challenges. Building upon these assessment findings, the adaptation policies that have been proposed or implemented by governments in the three major geopolitical blocks of our time (namely China, the European Union, and the United States under Biden administration) are then examined. Due to the distinctive economic, social, and political contexts of the decision-making structures in each of the corresponding administrations, many of the policies followed for resilience differ from each other. Nevertheless, by analysing these adaptation policies this chapter has identified common themes, reliable practices, and potential areas for improvement in developing resilient and sustainable transport systems globally. In this way it is able to give, in the concise form of a Table, typical policy instruments of each type that have been implemented in all three regions. Monitoring the success of the various adaptation policies is of crucial importance and this, heavily relies on the ability to quantitatively measure and monitor system resilience through well-defined indicators. Following this recognition, the chapter provides a brief overview of the various indicators that have been proposed and adopted to evaluate transportation system resilience, mainly focusing on topology-based, trip-based, and macroscopic fundamental diagram (MFD)-based indicators. These can reflect the real system resilience from a specific aspect or multiple aspects. At the end, an analysis is given of the applicability of the indicators of resilience in various types of disruptions and their integration with artificial intelligence (AI)-powered approaches. The material in this chapter, offers guidance in selecting appropriate resilience indicators and prioritising development investments, ensuring that resources are allocated effectively to enhance system performance and resilience. Lastly, the various AI applications for transport resilience are considered. The focus is on the current state and future advancements of AI-powered adaptation policies aiming to inspire stakeholders to re-evaluate the practical value and transformative potential of AI technologies in enhancing climate change adaptation strategies and redefining short-term response mechanisms.

This Chapter presents the AI technology and its dynamic role in promoting climate-friendly mobility. It first presents basic AI techniques that can perform tasks requiring human-like intelligence such as reasoning, decision-making, and learning. It then presents examples of specialized AI algorithms and models that are the backbone of artificial intelligence applications as they represent the mathematical and computational frameworks that enable machines to learn, reason, and solve problems. Short presentations are made to general application AI algorithms such as Supervised Learning, Unsupervised Learning, and Reinforcement Learning (RL) as well as to more specific ones such as for Natural Language Processing, Computer Vision, Robotics and Automation, big data handling and data mining. After a brief and concise presentation of the phases of developing AI software systems and platforms, the chapter presents and discusses various AI applications for reducing transport emissions. Of particular interest to the reader may be four specific AI applications that are given as examples from real-life systems and use cases that are presented and analyzed in length . These examples are presented in terms of the current state of technology, the role of AI in reducing carbon emissions, and their potential impacts. Finally, the chapter reviews the most pronounced challenges that exist for the widespread application of AI in climate friendly mobility such as its reliance on large-scale data, the need to construct robust network infrastructures for data transmission and real-time processing, the need for continuous iteration and optimization needed for AI systems, and the substantial computing and energy resources that are necessary.

This chapter investigates the use of Artificial Intelligence (AI) as an emerging and powerful tool in addressing the complex challenges imposed by adaptation of the transport sector to climate change. There is also a discussion on the potential of AI to help optimise the operation of the transport system in terms of traffic management, traffic safety, public transport operation, and the like, as well as in terms of aspects like social equity and accessibility, reducing mobility disparities, reducing carbon footprints and other. It comes as an additional examination of the issues of climate change adaptation discussed in Chap. 4 focused in more detail on the AI applications that can be used. As climate impacts intensify, the transportation system infrastructure as well as the operational elements of it will need to be protected against increasing risks from the impacts of extreme weather events. This chapter follows a holistic approach in presenting the need for a proactive climate adaptation while maintaining operational efficiency across the transport sector. It is based on the imperative to adapt transportation systems to increase resilience to climate change since failure to adapt and make those systems more resilient will result in large economic losses, grave safety risks, and disruptions to vital supply chains and mobility services. The chapter proposes an integrated framework for Artificial Intelligence applications for adaptation in the transport sector and discusses applications in predictive infrastructure maintenance, real-time monitoring and risk assessment, adaptive traffic management, optimised and inclusive transportation planning and design. It also makes detailed reference to the need for reducing mobility disparities, ensuring social equity and accessibility, establishing a viable workforce planning process, and to life-cycle analysis of mobility options.

This chapter deals with the issues regarding the utilisation of AI in the management and analysis of big data for climate friendly mobility applications. After a brief but concise description of the modern collection methods for such data it goes through the main AI techniques for data processing to ensure data quality. The main advantage of using AI for this sort of quality assurance exercise is its ability to train algorithmic models so that they can recognize trends and hidden issues and extract useful information with very short processing times hundreds or even thousands of times faster than traditional data processing methods. The text explains how machine learning or deep learning algorithms and other AI based technology can automate complex data analysis tasks, excel at identifying complex patterns and relationships from big data, discover valuable information hidden in these data and increase its accuracy and reliability. Consequently, there is a concise reference to the many AI tools and software libraries that can be used for such big data analytics. A few indicative examples of existing applications of big data in the transport and mobility sector follow as an indication of the use of AI techniques for this purpose in practice. With the help of AI, big data gets a number of quality attributes which include Accuracy, Availability, Integrity, Consistency, Completeness, and Precision. All these are explained and discussed in length. Finally, this chapter duels on the issue of energy consumption that is necessary for AI-assisted big data analytics and calls this situation “a mixed blessing”. This characterization is justified by the fact that whereas the use and utilisation of AI for big data analysis and handling has many advantages and can be a beneficial transformational force, it necessitates large quantities of electricity which if not generated by “clean” renewable sources it can generate more carbon emissions than saved from transport operations by using AI.

This chapter identifies some of the basic characteristics of human travel behaviour and investigates how AI can influence it towards meeting the strategic goals of sustainable and climate friendly mobility. It stresses the view that human travel behaviour and transportation system operation are two interlinked elements that interact in many and complex ways. Travel behaviour has been a major field of study in the Transportation science for many decades, and this chapter summarises many of the findings so far and explores ways in which AI can influence human behaviour to adapt and “match” the available transportation system capacity and make it more aware of the climate change and its impacts on transport and mobility. The chapter refers to the various Transportation Demand Management (TDM) policies (e.g., incentives, information dissemination campaigns, and other) and how AI can support their formulation and implementation. AI is seen as the catalyst for a whole new series of TDM models and policies so its potential application areas in this field are many and these are presented and discussed in detail. To support the view that when TDM policies are properly designed and consistently applied, they can significantly reduce vehicle travel even without the use of AI, the chapter refers to a recent report by the Australian Victoria Transport Institute where a series of data are presented showing the results of specific TDM policies that have been applied in the past in several urban areas. As TDM policies tend to have synergistic effects, their results are more pronounced if implemented as an integrated program that includes a combination of measures of different but complimentary nature. With the wider use of AI in TDM, it is reasonable to expect that the impacts will be higher, more widespread, and varied. Of particular interest is the list of application areas of AI in this field which are given in this chapter. These applications are aimed at influencing travel demand while at the same time optimising the transport system’s performance and operation.

The ethical implications of AI arise with respect to the moral values of humans and human societies. They are examined in this chapter under the three headings: Ethical implications related to the traveller, i.e. the decision making about travel (trip making) or transport (freight movement), Ethical implications related to the operation of the transport system, i.e., of the measures related to maximising efficiency and safety of the system and Ethical implications related to the vehicle, which refer to the use of the info and data about the driver of the vehicle. As regards the societal implications these relate to ways and means to safeguard that AI applications abide with commonly accepted social rules, conditions, and norms. The elements in each category of implications (ethical or societal) are discussed under names such as diversity, non-discrimination, fairness, privacy and security of users, and several others and then the various potential applications are classified under the following three categories: (a) Privacy, accountability, and security during the trip making. (b) Equitable use and accessibility of transport infrastructures and services, and (c) Maximum adherence to societal values, goals and objectives. As the need for ensuring equitable access to sustainable transportation is becoming a central issue of adaptation of the transport system in the face of climate change, this chapter further elaborates on the ability of Artificial Intelligence to support underserved and vulnerable communities in accessing climate-resilient transport services, to identify and support the mobility of vulnerable populations, to reduce mobility disparities, and to support inclusive transportation planning. At the end, the ethical implications of four specific mobility technologies are examined in more detail. These technologies are automated mobility, shared mobility, public transport, and freight transport and logistics.