**1. Introduction**

The current digital transformation offers substantial opportunities to industry for building competitive and innovative business models and complex circular supply chains; however, such a transformation also implies severe implications concerning sustainability, since the Information and Communications Technology (ICT) industry has a significant environmental footprint. In order to reach the milestones defined by the United Nations Agenda for Sustainable Development [1] and to implement the visions of circular economy, it is necessary to provide solutions in an efficient and sustainable way during their whole life cycle. Such a sustainable digital transition towards a smart circular economy is enabled by three key technologies: IoT, edge computing, and Artificial Intelligence (AI).

It is estimated that Internet of Things (IoT) and Industrial IoT (IIoT) technologies, which enable ubiquitous connectivity between physical devices, can add, only in industrial

**Citation:** Fraga-Lamas, P.; Lopes, S.I.; Fernández-Caramés, T.M. Green IoT and Edge AI as Key Technological Enablers for a Sustainable Digital Transition towards a Smart Circular Economy: An Industry 5.0 Use Case. *Sensors* **2021**, *21*, 5745. https:// doi.org/10.3390/s21175745

Academic Editors: Zihuai Lin and Wei Xiang

Received: 2 August 2021 Accepted: 23 August 2021 Published: 26 August 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

applications, USD 14 trillion of economic value to the global economy by 2030 [2]. In addition, the development of the classic view of the Internet of People (IoP) [3] and the Internet Protocol (IP) led to a convergence of IoT technologies over the last two decades, which paved the way for the so-called Internet of Everything (IoE) [4]. Such a concept is rooted in the union of people, things, processes, and data to enrich people's lives.

The explosion of IoT/IIoT technologies and their potential to pave the way to a more sustainable world (in terms of full control of the entire life cycle of products), can also lead to some pitfalls that represent a major risk in achieving the milestones defined by the UN Agenda for Sustainable Development [1]. As part of the IoT Guidelines for Sustainability that were addressed in 2018 by the World Economic Forum, a recommendation to adopt a framework based on the UN Sustainable Development Goals (SDGs) [1] to evaluate the potential impact and measure the results of the adoption of such recommendations was put forward [2]; however, in 2010–2019, and considering *Goal 12: Ensure sustainable consumption and production* [1], electronic waste grew by 38% and less than 20% has been recycled. Paradoxically, although these technologies have a huge potential for the digital transformation towards sustainability, they are not ye<sup>t</sup> contributing to the sustainable development of the ICT sector. Specifically, such a contribution is expected for the IoT sector, which has been seen as the driving force for a sustainable digital transition. The need for policies that effectively promote the sustainable development of new products and services is crucial and can be seen as a societal challenge in the years to come.

The concept of Green IoT (G-IoT) [5,6] is defined in [7] as: "*energy-efficient procedures (hardware or software) adopted by IoT technologies either to facilitate the reduction in the greenhouse effect of existing applications and services or to reduce the impact of the greenhouse effect of the IoT ecosystem itself* ". In the former case, the use of IoT technologies may help to reduce the greenhouse effect, whereas the latter focuses on the optimization of IoT greenhouse footprints. Moreover, the entire life cycle of a G-IoT system should focus on green design, green production, green utilization, and finally, green disposal/recycling, to have a neutral or very small impact on the environment [7].

IoT devices have increasingly higher computational power, are more affordable and more energy-efficient, which helps to sustain the progress of Moore's law to bring a sustainable IoT revolution in the global economy [8]; however, this vision directly collides with the advent of the concept of Edge Intelligence (EI) or Edge Artificial Intelligence (Edge-AI), where the processing of the IoT collected data is performed at the edge of the network, which imposes additional challenges in terms of latency, cybersecurity, and more specially, energy efficiency.

This article summarizes the most relevant emerging trends and research priorities for the development of Edge-AI G-IoT systems in the context of sustainability and circular economy. In particular, the following are the main contributions of the article:


The remainder of this article is structured as follows. Section 2 introduces the essential concepts that will be used in the article. Section 3 analyzes the main aspects related to the development of G-IoT systems, including their communications architecture and their main subsystems. Section 4 analyzes the convergence of AI and edge computing to create Edge-AI systems. Section 5 provides a cross-analysis to determine the key issues that arise when combining G-IoT and Edge-AI systems. Section 6 presents a practical Industry 5.0 application case and evaluates the energy consumption of a mist computing Edge-AI G-IoT model. Section 7 outlines the main future challenges that stand in the way of leveraging Edge-AI G-IoT systems. Finally, Section 8 is devoted to the conclusions.
