Is the energy management and building controls industry really implementing AI or is it just another buzzword?

Attending seminars, reading trade press and perusing tech brochures crammed with thought leadership pieces– what one buzz word pops up every time? Artificial Intelligence, or AI for the modern wordsmith.

I see more and more businesses adopting the phrase AI, stating they are AI ready.  But is the industry really delivering AI driven platforms or is it just another buzzword? First it was big data in the building controls and smart buildings scene, now it’s AI. Is AI the game changer or is it just innovation?

Artificial intelligence is termed as the simulation of human intelligence processes by machines and computer systems. These processes include learning (the acquisition of information and rules for using the information), reasoning (using rules to reach approximate or definite conclusions) and self-correction. AI is something that exhibits human-level reasoning, independence and creativity to ‘think’. 

A year ago I heard the phrase ‘data is more valuable than oil’ and it has stuck with me ever since because I believe it to be true. But the value doesn’t lie in the data; it’s what we do with it that counts. So is this where AI is making its mark?

We’ve nailed data collection in the energy and building controls sector as we’ve deployed IoT devices, developed and delivered cloud-based software and big-data platforms allowing more than 35% of organisations to rely on IoT for services.

However, it is what we do with that data that matters, the actionable insights derived from the vast quantity of data we collect, store and process is what will transform the services and solutions businesses design and deliver. Many say this is down to AI; but is it?

The challenge in turning that data into actionable insights, delivering to the business meaningful data upon which they can make business winning decisions for process improvements, energy efficiency, optimisation, work flow, wellness in buildings and even create new revenue streams by being able to design intelligent products and services requires analytical tools and human generated algorithms.

Whilst AI is predicted to be one of the areas that will shake up sustainability in 2019, with the technology being touted to be utilised for tasks ranging from heating management to transport route calculation; has the energy and building controls industry truly adopted AI yet or is it still just analytics and human generated algorithms – aka - machine learning?

Do the systems and technologies we are implementing for building and equipment control, energy efficiency and utilities management truly understand the business, building, and operational requirements?  Can they think for themselves, can they think like a human? With no human interaction what so ever? In my opinion, not yet.  

Data, analytics and machine learning are key to smart buildings and smart cities. IoT and analytical platforms are changing the way our buildings and cities work and how we interact with them. Analytics and machine learning enables systems to deliver insights and ultimately make ‘learned’ beneficial business or process change decisions and optimise operations and performance. Analytics and machine learning create a partnership between us and our buildings and systems and allow us to optimise our businesses, which benefit the bottom line. 

AI is promising new possibilities to streamline, expand and advance smart buildings and smart cities. I have no doubt that AI will ultimately transform every business in every industry, however if we look to the work by philosophers like Huber Dreyfus - a renowned Berkley philosopher – who was a critic of artificial intelligence since the 1960s, we find the true meaning of AI and why it may not be right here, right now for energy and building controls.

Huber Dreyfus argued that human intelligence and expertise depend primarily on unconscious processes rather than conscious manipulation, and that these unconscious skills can never be fully captured in formal rules.  

So is the industry really delivering AI or are we just machine learning? I agree with Dreyfus, ‘unconscious skills can never be fully captured in formal rules’.  
​
What we are seeing in our industry is sophisticated systems and devices that are able to collect, store and process a huge array of data sets at faster speeds than ever before. These systems and devices use statistic-based approaches to machine learning to stimulate the way the brain works by identifying anomalies and making quick decisions. ​

They are pushing data through human generated algorithms to provide a set of outputs from which decisions can be made and automated/programmatic actions undertaken.  

The systems are doing as we say, they’re not ‘thinking’ for themselves or using reasoning of their own making based on any knowledge of your building, business or the processes and people within them. Machine learning, in my opinion, is the closest we have come to making our systems and devices work like the human brain; not AI.