When Buildings Become Aware of Us

In recent years, we have witnessed an exponential grown in the volume and varieties of data, more powerful computational processing, and more accessible, affordable cloud data storage. These developments paved the way to produce models that are capable of analyzing bigger, more complex data, and deliver faster, more accurate results.

This may account for the growing rise in the adoption of AI in the built environment. Many consumers will have their first interaction with AI in the homes through voice interactions with smart home personal assistants by Amazon, Google, or Apple. 

Can we expect the same of AI in managing smart commercial facilities, going forward?

Why AI?

Smart cities and smart buildings generate an incredible amount of data daily. However, the data has not been systematically collected, stored, analyzed, or leveraged to boost efficiencies or to meet sustainability goals.

Here’s where we can harness the prowess of AI. Generally speaking, AI represents the broader concept of machines being able to carry out tasks in an intelligent way. Machine learning (ML), an application of AI, allows systems to automatically learn and improve from exposure to more data without being explicitly programmed. In other words, ML focuses on the development of computer programs that can access data and use it to learn for themselves. Such semi-supervised learning typically uses a large amount of input data but only a small amount of corresponding output data.

Working with AI/ML, critical information — such as where the problems are and what’s causing them — can be culled from the data flood using the appropriate algorithm, and delivered when and where such insights are needed. With such advanced tools, smart city planners and smart building owners will be better equipped to identify opportunities and to resolve problems. 

Can We Trust AI?

No doubt AI and ML hold enormous promise to improve our environment. But given these are still maturing as technologies, understandably many anxieties and concerns have been raised.

With ML, data quality and quantity are of paramount importance. Since a significant amount of data is needed to train the algorithm, a high quality of data input into the model is essential to enable better predictive capabilities. However, obtaining sufficient quality labeled data can be costly and often depends on human experts to perform the labeling.

We have to trust that the machine and data will make the right decision. AI and ML will start making sophisticated decisions as they become more advanced. One huge concern is if automated processes could “learn” patterns that lead to undesired or unintended consequences or biases. This is possible if the underlying data set included biases, or is collected from a process that structurally included some form of bias — the ML algorithms will replicate and perpetuate those biases.

Against this backdrop, business leaders must pay careful attention to the history and source of their datasets. Case in point: ML algorithms and the resulting models are created as combinations of numerous variables, and their predictions are not easily explainable. In such situations, users may find it difficult to trust the output of the algorithm. To smooth the transition, business leaders and policymakers need to be mindful of these concerns in enabling and applying "Responsible AI."

Putting ML in Smart Buildings

The next-generation smart buildings will be self-conscious, self-healing, and occupant-driven. Building occupants — such as employees, visitors, doctors, and patients — will be able to interact directly with their environment for better comfort and productivity.

The Bee’ah new headquarters in Sharjah, United Arab Emirates, is an example of an office of the future. Built with a wide array of intelligent edge systems, devices, and software, the building is designed to help improve occupants' productivity through a virtual AI persona to handle facility booking and ambient environment control, amongst others. 

ML is the foundation block for systems that rely on biometric recognition for controlled physical access. For example, cameras, which use supervised machine learning techniques like neural networks to identify users, can be used to verify users' identity with very high confidence. The combination of facial recognition and traditional card access (such as card or pin) provides a higher level of assurance for secure access while minimizing disruption for the users. The Bee’ah headquarters will be leveraging facial recognition technology to enable frictionless access or to control access to sensitive areas.

ML can also be applied for energy management and predictive energy optimization in commercial settings. Facility managers at the Bee-ah building, for instance, will have real-time visibility of all assets within the building. They can harness internal and external data to benchmark building performance, monitor building equipment, ensure occupant comfort, as well as forecast operational budgets.

Building owners can utilize ML and predictive analytics for better energy management. The AI-based technology can analyze load management, predict anomalies, and detect faults in various building systems such as heating, ventilation air-conditioning (HVAC), lighting, appliances, and devices. This allows potential issues to be addressed before something serious happens.

Predictive analytics can also optimize processes to manage the upcoming load. Take cooling, for example, the AI technology can intelligently pre-cool or store chilled water in cooling towers in advance of anticipated loads, or create just the right amount of chilled water early in the day before peak utility rates apply. The new Bee’ah building is designed to reduce water consumption by some 20 percent, is fully powered by renewable energy, and will have zero net energy consumption.

Based on feedback of human comfort and energy consumption, operational set-points for various components of HVAC systems can also be computed. With AI, the energy consumption of a whole building or even a piece of equipment can be predicted, which enables peak shaving, avoidance of utility penalty, and planning for cost-effective energy supply.

Voice control of building functions could be next for commercial buildings. Currently common in personal devices like smartphones, laptops, or wearables, increasingly voice control technology will be deployed in communal spaces, such as conference rooms. However, the take-up will largely depend on trust - such as the ease-of-use, how well the technology performs its intended use and the protection of users' privacy.

Future Lies with Intelligence

Without a doubt, data is a remarkable resource with fantastic potential; it will significantly benefit those who can gather it, understand its meanings, and put it to work. A data-driven approach towards smart facilities management is the way forward.

Such a strategy calls for two essential elements: the seamless integration of building systems and equipment to provide efficient, centralized control and using AI/ML to mine the data - generated by the linked systems and supplied by external sources - to discover opportunities for improved efficiency and performance.

Alvin Ng, vice president and general manager, Digital Solutions, APAC, Johnson Controls, wrote this article.

The views and opinions expressed in this article are those of the author and do not necessarily reflect those of CDOTrends.

Alvin Ng
Alvin Ng, Johnson Controls: For AI to be trusted, data quality and quantity are key.