Waste Management & Research A Stochastic Multi-agent System for IoT-enabled Waste Management in Smart Cities Theodoros Anagnostopoulos, Arkady Zaslavsky, Inna Sosunova, Petr Fedchenkov, Alexey Medvedev, Klimis Ntalianis, Christos Skourlas, Andrei Rybin and Sergei Khoruznikov Abstract Earth population is moving towards urban areas forming Smart Cities (SC). Waste management is a component of SCs. We consider a SC which contains a distribution of waste bins and a distribution of waste trucks located in the SC sectors. Bins and trucks are enabled with IoT sensors and actuators. Prior approaches focus mainly on the dynamic scheduling and routing issues emerged by IoT-enabled waste management. However, less research has been done in the area of stochastic reassignment process during the four seasons of the year over a period of two years. In this paper we aim to stochastically reassign trucks to collect waste from bins through time. We treat this problem with a multi-agent system for stochastic analyses. Keywords Smart Cities, IoT, Waste Management, Stochastic Analyses, Multi-agent System Introduction The vast amount of earth population (i.e., 70%) will move to urban areas by 2050, thus, forming vast cities United Nations (2014). These cities require a smart infrastructure to manage citizens’ needs and offer fundamental and more advanced services. Future Internet technologies enhanced by the use of the Internet Protocol (IP) on numerous wireless sensors will be adopted to enable the Internet of Things (IoT) paradigm. Plethora of sensors have the opportunity to be part of Wireless Sensor Networks (WSNs). When WSNs are applied in a city, they are responsible for collecting and processing ambient information and, thus, to upgrade legacy city infrastructure to the so- called Smart Cities (SCs).A definition of the concept of SC is provided in (Centre of Regional Science, 2007): “A Smart City is a city well performing in a forward- looking way in the following fundamental components (i.e., Smart Economy, Smart Mobility, Smart Environment, Smart People, Smart Living, and Smart Governance), built on the ‘smart’ combination of endowments and activities of self-decisive, independent and aware citizens”. This definition incorporates the fundamental component of a smart environment which is mainly adopted for systems dealing with environmental pollution. The concept of smart environments depicts the ambient intelligence found in a SC through the adoption of smart devices and wireless networks. This way, intelligent applications could be delivered on top of such infrastructures. WSNs are capable of reforming activities in a SC in every aspect of daily life. In this paper, we focus on a specific application domain, waste management. Waste management is treated under the umbrella of an intelligent multi-agent model to impact on the quality of life of citizens. The reason is that waste disposal has a clear connection with negative impacts in the environment and thus on citizens’ health. The SC we experiment with in this study is the city of St. Petersburg, Russia. St. Petersburg is a city of 5 million citizens covering a total area of 1,439 square kilometers, a density of 3,391 citizens per square kilometer. On average,solid waste produced in the city is 1.7 million tonnes per year. The daily amount of municipal solid waste generated is 0.93 kilograms per citizen. On a daily basis, the municipality of St. Petersburg uses 476 waste collection trucks with a capacity of 5 tons per truck. The fuel consumed in one year is,on average,1.8 million liters. The average costs spent for fuel in one year for waste collection is more than 1 million US dollars. Finally, the traffic congestion caused by the fleet of waste collection trucks at rush hours is significant due to the narrow roads and small backyards,causing indirect problems in citizens’ activities.Obviously, it is critical to efficiently manage the waste disposed in every location of a SC not only focusing on the collection activities but also on its transport and recycling. Waste management is treated as a set of services on top of an IoT infrastructure in a SC. These services cover the following parts of a waste management scheme: (i) waste collection planning and implementation (e.g., routing solutions for collection trucks, dynamic adaptation of routes), Corresponding author: Theodoros Anagnostopoulos is with the University of the West of Attica, Greece. He is also with the ITMO University, Russia. Email: thanag@teiath.gr