Modeling and Simulation of a Smart Street Lighting System

Abstract

Smart cities comprise highly dynamic, complex and softwareintensive systems planned and created to cope with problems inherent to the increasing world population, such as need for life quality in cities, intense traffic, and sustainability. Owing to such complexity and dynamics, static notations, such as Unified Modeling Language (UML) and Systems Modeling Language (SysML), are often unable to support a precise conception, planning and design of those smart cities. In turn, the combination of traditional modeling and simulation (M&S) could leverage the design of such systems by enabling the analysis of both structure and behavior aspects still at design-time. The main contribution of this paper is offering evidence to support the premise of M&S be a promising paradigm for software engineering, including in smart cities domain. We report findings on a study carried out for assessing, via M&S, two architectures for a Public Street Lighting System (PSLS) of a smart city. The first architecture uses a hub as intermediator between luminaries and PSLS. The second architecture uses software-intensive luminaries connected to the PSLS using publish/subscribe architectural style. Preliminary results show that the publish/subscribe architecture delivers a greater degree of correctness than the former. Additionally, we conjecture that the adoption of M&S can foster the identification of smart cities behaviors and the identification of, at design-time, properties that could be noticed only after the system has been already deployed.