Pannee SuanpangPitchaya JamjuntrPhuripoj KaewyongChawalin NiamsornKittisak Jermsittiparsert2025-03-102025-03-102023Sustainability (Switzerland)2071105010.3390/su150104552-s2.0-85146002255https://repository.dusit.ac.th//handle/123456789/4584The world is entering an era of awareness of the preservation of natural energy sustainability. Therefore, electric vehicles (EVs) have become a popular alternative in todayÕs transportation system as they have zero emissions, save energy, and reduce pollution. One of the most significant problems with EVs is an inadequate charging infrastructure and spatially and temporally uneven charging demands. As such, EV drivers in many large cities frequently struggle to find suitable charging locations. Furthermore, the recent emergence of deep reinforcement learning has shown great promise for improving the charging experience in a variety of ways over the long term. In this paper, a Spatio-Temporal Multi-Agent Reinforcement Learning (STMARL) (Master) framework is proposed for intelligently public-accessible charging stations, taking into account several long-term spatio-temporal parameters. When compared to a random selection recommendation system, the experimental results demonstrate that an STMARL (master) framework has a long-term goal of lowering the overall charging wait time (CWT), average charging price (CP), and charging failure rate (CFR) of EVs. © 2022 by the authors.All Open Access; Gold Open Accessdestinationelectric vehicleelectronic vehicle chargingintelligent recommendation systemsmart citysmart tourismArticleScopus