A house that knows you: creating smarter homes and services

"AI and IoT provide tremendous insights… inside of buildings."

By | 4 minute read | March 27, 2019

Courtesy of Huf Haus

We get to know our homes and their quirks—the radiator that pings and hisses as it overheats a room to tropical levels, the hallway floor that creaks like a haunted house, the shower that takes seemingly forever to warm up.

This relationship is intimate and emotional, and at times our homes seem almost alive.

“Buildings have long mimicked living organisms–plumbing circulates through the walls, wires innervate every room, and concrete and beams provide skeletal support, but until recently, buildings lacked the most critical body part: a brain,” said Stephan Biller, Vice President of Watson Internet of Things (IoT) at IBM.

When buildings have a “brain,” or become smart, they can serve our needs in new and unprecedented ways.

One example is Huf Haus, the German pre-fabricated home manufacturer. They have created the world’s first house that learns from its residents.

Called “Outlook,” the smart home gets to know people through their interactions, which are collected through a network of sensors and analyzed by IBM’s Watson IoT platform. Huf Haus also uses Watson Assistant to ensure the home and its occupants can interact naturally.

“Artificial intelligence and IoT technology provide a tremendous amount of insights to improve the lives inside of buildings,” said Srijnan Sanyal, Leader, Global Watson IoT Industry Lab. “This includes personalization, energy management, digital assistants, security, and more.”

Through voice commands, facial recognition* and understanding behavioral patterns, Huf Haus learns how to make its residents as safe and comfortable as possible and adapt to changing living or working situations.

Huf Haus will remember lighting and temperature preferences in the morning and how a resident likes the drapes open if it’s a sunny day. It will have their coffee ready and the news on—and even can recommend what to wear depending on the weather.

Forgetful residents? Huf Haus can remind them of a doctor’s appointment that day and suggest the best route to the appointment, based on traffic patterns and events. If a friend visits, the home will identify them via facial recognition* and unlock the door. It can sense if it’s noisy outside and close the windows.

This ultimate personalization even extends beyond the home, said Sanyal.

“If you are watching a Netflix show on your smartphone when traveling in an Uber, the show will continue to play on your TV when you get home,” said Sanyal. “And the digital assistant in your car will know to guide you to your doctor’s appointment, but will call into a work meeting on the way.”

All about that data

Today’s smart buildings collect massive amounts of data. And by 2022, 500 smart devices will be present in a typical family home—adding even more data.

This level of personalization in smart homes is made possible by integrating and converting data across devices into insights that help us better understand customers.

But personalization is just one prime example of how data can change our lives.

“By aggregating the data of millions of homes, we can learn how to better protect the privacy and ensure the security of homeowners,” said Sanyal. “Energy and utilities companies could better manage their gas, water, and electricity distribution to homes, neighborhoods, and cities.”

AI could also be infused into hospitality applications, say to serve as a smart assistant for hotels. It could have a future in retail, banking, and just about any industry, as it could learn a specific industry’s lingo.

For manufacturing, automotive, and electronics companies, this gold mine of data helps them to create new services and products for their customers.

“Automotive companies are collecting enormous amounts of data and can use that to offer new services in a non-intrusive way, like traffic information, restaurant or grocery store recommendations, or even better insurance,” said Sanyal.

Companies can go beyond individual sales transactions to build a lifelong relationship. One McKinsey analysis across 30 industries showed that average earnings-before-interest-and-taxes margin for aftermarket services was 25 percent, compared to 10 percent for new equipment.

Office buildings can also be made smarter, benefiting not just building management but also the office residents.

KONE, a global leader in the elevator and escalator industry, is already analyzing vast amounts of data from sensors embedded in elevators to identify and predict maintenance issues.

KONE’s products and services blend cutting-edge technology with an understanding of people’s needs to improve people flow in homes, offices, neighborhoods, and cities. In both residential and office buildings, different stakeholder groups—developers, facility managers, building residents—are taken into account, as each have unique challenges and needs.

As with Huf Haus, the goal for every industry is to open up new possibilities for helping people, their customers.

“All this data provides new revenue streams and value for industrial companies,” said Sanyal. “In the future, it will open up even more powerful opportunities for changing and integrating society and our world.”

*As of June 2020, IBM no longer offers general purpose IBM facial recognition or analysis software. To learn more, please see IBM CEO Arvind Krishna’s Letter to Congress on Racial Justice Reform [https://www.ibm.com/blogs/policy/facial-recognition-sunset-racial-justice-reforms/].