ASAS: An Approach to Support Simulation of Smart Systems

Abstract

Smart systems, such as smart cities, smart buildings, and autonomous cars, have recently gained increasing popularity. Each such system is essentially a System-of-Systems (SoS). SoS are dynamically established as alliances among independent and heterogeneous software systems to offer complex functionalities as a result of constituents interoperability. An SoS often supports critical application domains, and, as such, must be reliable. Many SoS have been specified and evaluated for their correct operation using static models. However, specification languages have not supported to capture their inherent dynamic nature nor enabled to monitor their operation. The main contribution of this paper is to present ASAS, an approach to Automatically generate Simulation models for smArt Systems (ASAS) in order to support evaluation of their operation. In particular, our approach makes it possible to transform formal models of the SoS architecture (expressed in SoSADL) into simulation models (expressed in DEVS). We evaluated our approach by conducting two case studies using a flood monitoring system that is intended to be part of a smart city. Results indicate that ASAS can successfully generate functional simulations for the SoS operation, which in turn can enable to reason and monitor an SoS operation, taking into account its dynamic nature.

Simulation of Systems-of-Systems Dynamic Architectures

“…In our approach, dynamic reconfiguration in an SoS software architecture is based on constituent addition, constituent removal, constituent replacement, and architecture reorganization. Our approach uses a SosADL to DEVS model transformation (SosADL2DEVS) that automatically generates simulation models from SoS software architecture specifications, based on an approach called ASAS (Approach to Support Simulation of SmArt Systems) (Graciano Neto et al 2018a). However, ASAS does not support the simulation of dynamic architectures.…”

“…However, ASAS does not support the simulation of dynamic architectures. Dynamic-SoS was established by adding the generation and monitoring of dynamic architectures for that SosADL2DEVS transformation produced by Graciano Neto et al (2018a), as depicted in Figure 1. Moreover, we established a set of activities to systematize the use of Dynamic-SoS approach, as follows.…”

Architecting exogenous software‐intensive systems‐of‐systems on the internet‐of‐vehicles with SosADL

“…It takes as input the concrete SoS architecture description, the operational environment description (via concrete specifications of environmental gates given how constituent systems sense local environments), and produces different simulation models in Discrete Event System Specification (DEVS, http://www.ms4systems.com/): one atomic model for each constituent system and one for each mediator, as well as a coupled model representing the coalition of mediated constituent systems. The SoS architecture validator was implemented as a distributed simulator for SoS architecture. The approach for generating the SoS simulation models is based on the translation from SosADL to DEVS, via model‐to‐model transformation, and then from model‐to‐text for transforming DEVS into DEVSNL, the input language of the MS4Me simulator for DEVS.…”

Simulating Systems-of-Systems Dynamic Architectures

“…Our approach uses a SosADL to DEVS model transformation (SosADL2DEVS) that automatically generates simulation models from SoS software architecture specifications, based on an approach called ASAS (Approach to Support Simulation of SmArt Systems) (Graciano Neto et al 2018a). Dynamic-SoS was established by adding the generation and monitoring of dynamic architectures for that SosADL2DEVS transformation produced by Graciano Neto et al (2018a), as depicted in Figure 1. Moreover, we established a set of activities to systematize the use of Dynamic-SoS approach, as follows.…”

“…(2017). Simulação In addition to these two self-authored important publications, other several publications were achieved using results provided by the approach developed by this student and having him as a co-author, including one journal publication Qualis B1 (Graciano Neto et al 2017), two full papers in premier international conferences Qualis A1 (Graciano Neto et al 2018c;Graciano Neto et al 2018a), and also other important four conferences and workshop papers 4 .…”