| Abstract |
Effective quality management plays a pivotal role in ensuring the smooth operation of smart city systems, which have significant implications for safety, accessibility, affordability, and maintainability. Dependability of autonomous systems is of utmost importance, as achieving satisfactory levels of availability and reliability poses considerable challenges. Smart cities are characterized by interconnected sub-architectures, encompassing vehicle monitoring, sidewalk monitoring, and building monitoring, all of which need to function efficiently. Analytical models such as Petri nets, Markov chains, and fault trees are well-suited for evaluating complex scenarios in the context of smart cities. This paper presents analytical models that utilize fault tree and Markov chain techniques to assess the availability and reliability of smart city monitoring systems. The model is divided into shared and non-shared components, with non-shared components being specific to certain contextual applications, while shared components, such as data processing and electrical power, are essential for all smart city monitoring and management systems. The study underscores the ease with which the fault tree model can enhance availability by modifying failure requirements and resources. Case studies provide concrete examples of how availability improved from 95.3 to 99.8% by varying a configuration known as "KooN" in multiple components. This paper takes a comprehensive approach to evaluating the dependability of smart city architectures and contributes advancements, such as hierarchical modeling, sequential sensitivity analysis, and the "KooN" analytic method. These contributions expand the existing knowledge and methodologies in smart city dependability analysis. Moreover, this work aims to serve as a practical tool to assist smart city managers in optimizing their proposals. All modeling aspects and parameters are detailed thoroughly to enable effective implementation of the proposed approach by anyone using it. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024. |
