Cloud vs edge AI: What's best for your facility?

Building managers are integrating smart technology into the properties they are responsible for at an unprecedented pace. According to Juniper Research, the number of smart buildings worldwide will grow 150% by 2026, from 45 million buildings this year to more than 115 million. There are good reasons for this dramatic increase in deployment. Cutting-edge automation software and systems offer owners the opportunity to continuously monitor operational parameters such as occupancy, indoor air quality (IAQ) and utility usage to help achieve unprecedented levels of safety and efficiency.

However, integrating smart technology into facilities can be unsettling for some building managers. The decisions that must be made when adopting automated systems are complex and may include elements with which they are unfamiliar. But just as they have mastered HVAC, lighting controls and chillers, building managers can also learn about the Internet of Things (IoT), networking and artificial intelligence (AI).

Artificial Intelligence-enabled Internet of Things (AIoT) systems can be particularly intimidating, but they can be one of the most powerful ways to maximize building efficiency, safety, and sustainability. AI can be applied at the edge (Edge AI) or the cloud (cloud AI). Both have their advantages, depending on the goals and needs of the application, and construction managers who know when to use each (or a combination of both) have an advantage.

Understand the difference between remote storage and local storage

AI being deployed today was originally born as a cloud computing technology. The machine learning algorithms behind these systems require significant computing power, both to train the algorithms and to invoke them to provide insights—a process called inference. Until recently, local infrastructure rarely had the resources to do these things efficiently, so building operators had to run their AI applications outside of data centers.

However, running smart building applications outside of remote data centers has its own limitations. Connectivity, bandwidth costs, security and latency – the time it takes to send data to the cloud and back – can impact a system’s efficiency. If a machine or the entire building automation system is about to fail, alarms and automated responses need to be as immediate as possible.

A new generation of edge computing technologies alleviates this problem to a large extent: the infrastructure installed in facilities has the processing power required for these compute-intensive workloads.

Companies like FogHorn, founded seven years ago, have developed edge artificial intelligence technology that creates new possibilities for digitally transforming building operations. This includes advanced techniques (called Edgification) for optimizing AI models to run efficiently on low-cost edge computing devices. Johnson Controls acquired FogHorn in early 2022 and has now integrated edge technology into its OpenBlue platform.

By bridging the local capability gap, edge devices provide an architectural component important to the goal of running a building as efficiently and effectively as possible.

Choosing Between the Cloud and Edge AI

With the advent of Edge AI, construction managers considering implementing intelligent automation technology are now almost inevitably faced with whether to deploy on-premises or The problem of deploying AI in the cloud. For those facing this problem, there are some simple rules of thumb to consider.

Edge AI works best when:

• Needs to perform operations in real-time or near real-time. Intelligent automation systems that detect operational issues and automatically issue alerts or responses tend to work best when delays are minimized.
• Requires local control of the system. Shutting down machines or adjusting control systems from the cloud often comes with security and latency challenges.
• There are limitations on data transfer and storage costs. Consider, for example, a video surveillance system where high-fidelity images from multiple cameras are analyzed by a computer vision AI model, a popular AI application. Sending all your data to and storing it in the cloud can quickly become expensive.

The cloud may be better if:

• Complete rigorous data analysis. Construction managers often want to gain a deeper understanding of how they operate based on AI analytics, or run simulation exercises on “digital twin” versions of their facilities. This kind of data analysis typically doesn't need to happen in real time, so it's best performed in the cloud, where managers can leverage the most powerful hardware and software tools to do the job at any scale.

A combination of the two may be best in the following situations:

• Running multiple buildings and correlating information between them. The cloud allows for a centralized data clearinghouse and command center. In practice, a hybrid approach is often adopted, where some initial processing in individual buildings occurs via edge AI, and then cloud AI is run on aggregated data from multiple buildings, possibly combined with other data sources.

Take Your First Steps to Adopting Artificial Intelligence

It’s important to remember that these are decisions that construction managers don’t need to make alone – there are expert technology providers You can ensure that AI is deployed where it best meets your unique needs. Construction managers don’t need to be data scientists and fully understand all aspects of AI and its underlying machine learning algorithms, but can instead work with specialized technology vendors to let AI work its magic behind the scenes.

Oracle, like many organizations now embarking on mass return-to-work policies, views the aftermath of the pandemic as a unique moment to introduce smart building systems. After several years of pandemic-related closures, employees are clinging to physical workplaces where amenities are at their fingertips, collaboration tools are ubiquitous, air quality is monitored, crowding is limited, and their companies are in the energy Sustainable development goals are achieved in terms of energy and energy use, water and waste reduction. With building occupancy rates still at historically low levels, shutting down systems that don’t need to be running can help significantly increase efficiency.

These changing workplace dynamics and expectations can be an opportunity to evaluate new investments in IoT technologies, the advanced networks that connect them, and the artificial intelligence systems that control them. This is also an opportunity to evolve the workplace to make better decisions based on occupancy, employee experience needs, site ownership and its use case criticality (e.g. research lab vs. office space).

Construction managers have historically prioritized schedules when deciding whether to invest in automated control systems. no longer. The new key consideration is utilization metrics. They can't take it for granted that everyone will come back, and many companies are adopting hybrid work policies.

Creating Smarter, Safer and More Sustainable Spaces

For the first time, the office needs to compete with the home as an attractive and productive work environment. People want to feel confident knowing that their office's indoor air quality (IAQ) is monitored, resources like water and energy are used efficiently, and the rooms they live in are comfortable. AIoT systems can help make buildings more energy efficient, healthier, more autonomous, safer, and more responsive to the needs of their occupants.

In response, building managers new and old are seeking support from smart technology providers to help them acquire the new skills needed to implement AIoT automation systems and optimize operations. A valuable lesson is when to deploy AI on-premises or in the cloud. Once they determine whether edge or cloud AI aligns with their building goals and application needs, informed building managers can trust that AI will help them ensure healthy air, comfortable spaces, and efficient operations to help revitalize their buildings.

The above is the detailed content of Cloud vs edge AI: What's best for your facility?. For more information, please follow other related articles on the PHP Chinese website!