The past decade has seen a rapid penetration of electric vehicles (EVs) as more and more logistics and transportation companies start to deploy electric vehicles (EVs) for service provision. In order to model the operations of a commercial EV fleet, we utilize the EV routing problem with time windows (EVRPTW). In this paper, we propose an end-to-end deep reinforcement learning framework to solve the EVRPTW. In particular, we develop an attention model incorporating the pointer network and a graph embedding layer to parameterize a stochastic policy for solving the EVRPTW. The model is then trained using policy gradient with rollout baseline. Our numerical studies show that the proposed model is able to efficiently solve EVRPTW instances of large sizes that are not solvable with current existing approaches.

Millions of people have limited or no access to electrical energy. A diverse group of scholars and activists, from individual researchers to nongovernmental organizations (NGOs), civil society organizations, UN, and increasingly to engineering and financial private sector firms, have been working on initiatives to address this issue, which is considered by many a human rights problem. Energy access is not only a challenge for developing economies but is also equally important for service to remote areas such as the Arctic, islands, and communities distant from the grid. Several governments and private institutions, as well as nongovernment organizations around the world, are now funding projects and a variety of initiatives to help impoverished communities, especially across Africa and remote regions, develop clean and sustainable energy systems. The IEEE is actively involved in addressing energy access issues through several initiatives, such as the Smart Village program, focusing on “integrating sustainable electricity, education, and entrepreneurial solutions to empower off-grid communities” (http://ieee-smart-village.org/). In this special issue, we concentrate on highlighting the current state of knowledge associated with strategies for bringing clean, affordable, and sustainable electricity service reliably to energy-limited communities, based on local renewable energy (RE) sources and storage systems. Sociopolitical and economic forces as well as inclusionary and exclusionary culture or practices that often define the technical approaches and solutions offer opportunities for progress on this important problem.

Bold actions are necessary to unlock the potential for economic empowerment by eradicating energy poverty (UN Sustainable Development Goal 7) by 2030. This will require a sustained commitment to significant levels of new investments. Delivering on the promise of universal energy access and improved life quality has eluded policy-makers and governments over the past seven decades. Affordability of energy services for every global citizen, spanning vastly diverse regions and local contexts, requires the development and massive diffusion of technologies that offer “point-of-use” options combined with new business models. Social innovations and flexible governance approaches will also need to be integrated with technological advances. The scope and scale of developmental change span large-scale grid systems to decentralized distributed resources at community levels to the households. We recommend a global network of “energy access innovation centers” dedicated to providing a dynamic “extension service” that bolsters the entire supply chain of talent and expertise, design and operational requirements of system deployment and capacity to embed low-cost, high-performance next-generation technological solutions in the field. To meet the needs of those at the base of the economic and social pyramid, the dual challenges of economic development and transition to a low-carbon energy future make clean energy access the quintessential challenge of the 21st century.