The continuous migration of people towards cities has led to an increase in resource consumption among homes, primarily in terms of energy and water. This phenomenon calls for resource management solutions in an attempt to drive sustainable consumption patterns. Among these solutions lie Home Automation Systems (HAS) capable of monitoring and controlling different appliances so as to deliver services such as appliance control and security surveillance. As in most HAS architectures today, these appliances are uniquely identified and connected to the Internet, as in Internet of Things (IoT) networks. Though, despite their potential, such architectures generally fail to address four essential features altogether: easy adaptability, access to remote services, interoperability and software portability. In this work, we propose a reference HAS architecture which implements the previous features. Our approach is based on the synergy between the FIWARE IoT middleware and a multi-agent system (MAS), leading to an autonomous IoT (AIoT) system. To demonstrate the applicability of this architecture and to evaluate its potential to drive sustainable resource consumption, we have deployed a Home Energy Management System (HEMS) in accordance with the proposed architecture, and gathered results of various simulations of a home environment managed by the HEMS. Our results suggest that the HEMS can aid home-owners in decision making, raise awareness as of their resource consumption profiles, and provoke behavioral changes leading to more sustainable consumption patterns.

The concept of fairness has been studied in philosophy and economics for thousands of years, so human actors in social systems have had plenty of time to “learn” what does, and does not, work. Yet, only recently. However, it is a relatively new question how software agents in a multi-agent system can use Reinforcement Learning models to develop an architecture that promotes equality or equity in the distribution of rewards to the agents within the system. Recent significant contributions have focused on optimising for efficiency based on the assumption that efficiency and fairness are opposites to be traded off against each other, but actually, the result of mixing fair and efficient policies is unknown in multi-agent reinforcement learning settings. In this work, we experiment with fair and efficient behaviours jointly, based on an extension of the state-of-the-art model in fairness SOTO that intertwines efficient and equitable recommendations. We analyse the fair versus efficient behavioural spectrum in the Matthew Effect and Traffic Light Control problems, finding some solutions that outperform the baseline SOTO and others that outperform a selfish baseline with comparable architectural design. We conclude it is possible to optimise for fairness and efficiency and this is important when computation of the reward distribution has to be paid for from the rewards themselves.