While there are several partial solutions to model some aspects of cities (e.g. transportation or energy), there is no framework allowing modelling of a complex system such as a city. This paper aims on providing a solution that can be used by practitioners to model impact of different scenarios and smart city projects encapsulating different subsystems, such as transportation, energetics or, for example, eGovernment. The term “smart cities” is classified into Systems Theory, particularly focusing on Cyber-Physical Systems. This classification is further elaborated to define a new term, so-called Smart City Agent (SCA). The SCA is considered as the main building block for modelling smart cities. The approach within this paper however stresses the interconnection of different systems within a city. Its’ strength is in better exchange of data and among heterogeneous agents. This information management approach is the missing key in the growing market of partial smart city solutions as it will allow simulation of solutions in complex systems such as a city. The suitability of usefulness of the proposed approach is demonstrated on a use case dealing with charging of electrical vehicles. The results show that the approach is suitable for modelling of dynamic behaviour.

Introduction

The area of smart cities is currently undergoing a quick development and many different solutions are emerging on the markets. It is estimated that by 2030 more than 100 billion dollars will be invested in smart city applications (Visvizi & Lytras, 2018). In the paper (Lom & Pribyl, 2017), a modelling approach of smart cities called SMArt City Evaluation Framework (SMACEF) was introduced. SMACEF is the modular framework that allows modelling of the current state of a system as well as its future states, and based on defined scenarios and key performance indicators (KPIs) can be benchmarked which the proposed solution is the best one. In other words, the goal of the framework is to evaluate if the proposed solutions are beneficial and useful for cities or not. The modelling approach is based on the Multi-Agent Systems (MAS). Every object is represented by an intelligent agent. The practical implementation of SMACEF was published in the paper (Pribyl, Lom, & Přibyl, 2017). Based on SMACEF, a new type of an intelligent agent – Smart City Agent (SCA) is introduced and described as a building block for modelling smart cities in this paper. The area of smart city modelling is classified to the theories of Systems Theory and Cyber-Physical Systems, and both can be modelled by Multi-Agent Systems. The Smart City Agent is a modified version of an intelligent agent. It is more suitable for benchmarking and evaluating purposes in smart cities. The practical example of an implementation of Smart City Agents using SMACEF is demonstrated. Cities are dynamic and nonlinear systems and for this reason the models have to be dynamically simulated with different scenarios, and the results of these simulations should be benchmarked.