In the early days of cloud computing, many experts predicted a massive transfer of industrial data and operations to the cloud. This mindset has gradually changed as organizations and cloud service providers realized that it is better to use the cloud only when its scalability, capacity and quality of services are needed, while using services close to the field in other cases such as when real-time data processing is required. Recent studies anticipate that most industrial data will be stored, collected and managed closer to the field data sources rather than within centralized cloud infrastructures.
Therefore, edge computing is gaining significant traction when it comes to developing, deploying and operating applications in industrial environments, including applications for modern cities that are developed for smart buildings and other types of public infrastructure. Edge computing takes a decentralized approach to collecting, managing and analyzing data, which moves computing resources closer to the edge of the networking infrastructure and to the sources of data.
This distributed computing model can allow for faster data processing, reduced latency, and improved reliability, while at the same time delivering security and data protection benefits. By bringing computing power closer to the data source and reducing latency, edge computing enables real-time monitoring and analysis of critical data, which is already revolutionizing various applications, including smart cities and buildings maintenance applications. Moreover, edge computing is increasingly combined with embedded machine learning paradigms, which enable the execution of advanced Artificial Intelligence (AI) applications at a city’s or a building’s network edge. Read more about the current status of Edge AI in our comprehensive report.
Modern smart cities deploy, operate, and manage a considerable number of physical assets, which enable the operation of their critical infrastructures such as energy management and water management infrastructures. This is also the case for modern buildings and real-estate facilities, which comprise and manage many different assets such as building automation systems, HVAC (Heating Ventilation and Air Conditioning) units, smart meters, sensors and various other IoT (Internet of Things) devices. Therefore, both cities and facilities managers are putting an emphasis on the development and deployment of effective asset maintenance applications. The latter are important to ensuring the proper functioning and the longevity of critical infrastructure assets, while at the same time impacting the cost-effectiveness of critical operations.
State-of-the-art asset maintenance applications optimize the performance and reliability of physical assets by leveraging sensors, IoT and data analytics technologies. As a prominent example, sensors enable the collection of data about the actual condition of the assets towards implementing condition-based monitoring and condition-based maintenance (CBM) approaches that optimize the Overall Equipment Efficiency (OEE). Rather than implementing preventive maintenance approaches that replace or service assets at fixed timeslots in advance of their nominal End of Life (EoL), CBM moves service and replace times much closer to the assets’ actual EoL.
Effective asset maintenance applications can be used to predict and anticipate asset failures towards developing optimal maintenance schedules. The latter help cities and facility managers to avoid costly emergency repairs and equipment failures. More generally, predictive maintenance approaches for physical assets can increase the overall reliability of the critical infrastructure, while optimizing the cost-effectiveness of the maintenance processes. Furthermore, optimal maintenance schedules are also used to improve the allocation of resources (e.g., maintenance engineers, field workers, maintenance tools) to various asset management processes.
Most importantly, maintaining assets in a good working condition is essential for the safety and satisfaction of residents and visitors in smart cities and smart buildings. Specifically, asset maintenance applications can help the cities to promptly identify and address potential safety hazards. This can greatly reduce the risk of accidents and improve the overall quality of life for city dwellers. Likewise, citizens value urban environments where buildings and critical infrastructures are properly maintained.
Nowadays, many asset maintenance applications are cloud-based. These cloud applications collect, process and analyze large volumes of data about the physical assets in support of CBM and predictive maintenance applications. Nevertheless, this cloud-based approach to asset maintenance is gradually changing as edge analytics and edge ML enable value-added functionalities that are not possible without edge computing. In particular, edge analytics and edge ML offer the following benefits:
Overall, edge computing makes it possible to analyze complex datasets locally, which obviates the need for constant data transfer to a central cloud. This enables fast and efficient analysis, which allows for the detection of patterns, anomalies, and trends in real-time. Hence, the application of advanced analytics and Machine Learning (ML) models at the edge facilitates the real-time prediction of failures and maintenance needs, while at the same time boosting the production of optimized maintenance schedules.
Edge machine learning provides limitless opportunities for innovative asset maintenance applications in smart cities. These opportunities span different areas of the smart city’s infrastructures, including lighting, energy, transport, and water management infrastructures. Some of the most characteristic applications include:
Some of the above-listed edge functionalities for cities infrastructures maintenance are directly applicable to smart buildings as well. This is for example the case of the lighting maintenance and water leaks detection functionalities. Nevertheless, smart buildings have their own edge analytics and edge machine learning use cases that can save costs for facility managers, while improving the comfort and satisfaction of the tenants. Here are some prominent examples:
As cities continue to grow and evolve, the importance of effective maintenance strategies becomes paramount. Edge computing, edge analytics and edge machine learning offer exciting opportunities that revolutionize city maintenance practices by bringing advanced analytics and intelligent decision-making capabilities closer to the data source.
From real-time monitoring and anomaly detection to predictive maintenance and traffic optimization, the applications of edge analytics and edge machine learning in city and building maintenance are diverse and far-reaching. Modern cities must therefore embrace this powerful combination of technologies to enhance the efficiency, reliability, and sustainability of their critical infrastructures, but also to ensure a safer and more resilient urban environment for their citizens.
In the coming years, demand for edge functionalities is expected to continue to grow as cities and buildings become increasingly interconnected and data-driven. Edge functionalities are expected to proliferate due to the evolution of available networking infrastructures (e.g., the advent of 5G and 6G network) networks, the proliferation of IoT devices, and the wider availability of ML functionalities at the edge. The latter will enable more advanced and intelligent maintenance applications.
Overall, edge computing holds immense potential for transforming city and building maintenance practices. With a forward-thinking approach, cities and building owners can unlock the full benefits of edge computing in the maintenance of their critical assets.