How to Improve People’s Communication with Smart Buildings
The main objective of a smart building should be to help people to do the things they are trying to do, in the most effective way. Currently, smart buildings and the way users need to communicate with them, can be a source of frustration. There are some practical ways to help people and buildings play nicely together, from better design to voice control.
For some time, buildings and offices have been run using technology; systems that keep the environment comfortable and ensure safety and security. Over the last 5 years, conversations have turned to smart buildings. In fact ‘smart’ is now added to anything from speakers to forks!
Hype aside, smart buildings can create better places to spend time and give a more personalised experience for users.
Done badly, they can also be a source of frustration. In the past, flicking a light switch or turning a thermostat dial would be the most communication a user would have with a building. Now someone might have to interact with multiple pieces of hardware and software to get things done throughout the day. I believe we can improve this human-building interaction through both through passive methods, which reduce the need for direct communication between people and buildings, and active measures which improve communication that is still needed, making it more natural and intuitive.
Smart Building Components
The underlying systems in a building have not changed dramatically in their function, with a few main components. Namely, heating (plus cooling and ventilation), lighting (including blinds) and access control. Systems like fire detection and security play an important but often invisible role too. Mostly, when people say smart building, they are really talking about integrating and automating the different systems in a building and providing a better user interface.
In traditional buildings, these systems ran independently of users; they tended to either fail to keep people comfortable or use lots of energy by running systems to a fixed schedule regardless of demand, or both. We improved on this when we started to use sensors to adjust buildings based on what was happening inside them, looking at temperature, humidity, CO2 and smoke, plus other factors such as occupancy, scheduling and external weather data.
Globally, employee comfort and well-being is a higher priority than cost efficiency in real estate strategies, for the second year, according to the Verdantix Global Survey 2019. But users shouldn’t need to be involved in achieving this; the environment should simply meet people’s requirements automatically. Room level integration helps, by considering the linked effects of different systems, like opening the blinds when there is no direct sun and turning down the lights. One simple controller on the wall is also preferable to three separate switches and a needlessly complicated air-con remote; the kind where you press every button and eventually something happens, but nobody knows why or how to do it again.
The Internet of things (or IoT) brings a mass of relatively low cost, internet connected sensors that allow us to measure more parameters with greater granularity within a building. We can easily measure how many people are using a room, how the air quality and light levels vary throughout, which windows are open and even if the coffee machine has run out of milk. Combining this greater awareness with more intelligent control, the building can better satisfy the people it houses.
Everyone feels temperature differently and personal preferences vary. It is important to take this into account when setting the desired levels to avoid thermostat wars. Using Indoor positioning it is possible to automatically determine the best levels, based on the saved preferences of all people present in a space. We can help people to help themselves; by showing visually how temperature varies between rooms and across open-plan floors, people can chose to move to a spot that better suits their preferences. People like to feel in control, so we can give employees the option to adjust the temperature or lighting in the area they are in. When people disagree and put in different requests, systems can determine the best middle ground and show users what is happening.
People think that they hate the temperature in the office, but what they’re really upset about is a total lack of controlComfy Blog
Looking beyond comfort, flexible spaces are a trend within commercial environments and the right technology will be needed to make this work. Modular commissioning means that spaces can be changed to suit different uses, think moving walls and furniture options, so a large conference room can become three small rooms. The systems can adapt to the new setup without recommissioning (or having to go next door to wave at the motion sensor).
Deciding what space to provide can also be supported using IoT sensor data and clever analysis. By measuring how may people use rooms, areas or desks, and how often, means that building managers can see at a glance if there is a shortage or surplus of different spaces.
When people do need to communicate with smart buildings, we should design interfaces that make this simple and intuitive. User interfaces need to do more than just provide access to functions. These interfaces should suit the type of user and be situational; give people access to just the functions they need, at the relevant moment. I am unlikely to need to know what car parking spaces are free when I’m sat at my desk. Similarly, a maintenance engineer may need detailed asset information and live data while they are responding to a fault with the heating system, which would not be relevant at other times or to other people.
The number of buttons, apps, portals and technical know-how needed to get things done should be limited. A single point to control all tasks makes life quicker and easier for users, standardising this allows users to become familiar. In many buildings the reception desk becomes this single point, which is costly and probably annoying for people involved. There are great examples of workplace applications providing just this; a way for employees to find places and information, book rooms and desks and report things that are broken. Expanding this kind of app to cover all the common tasks in the workplace could add up to lots of time saved, and more importantly reduce frustration.
Next Step, Voice Control?
Home automation has been leading the way when it comes to user experience. Digital assistants, such as Alexa and Google Home, have brought natural speech recognition to our homes. Natural speech algorithms are, thankfully, getting better at recognising what humans say; Alexa ‘answered queries accurately 80% of the time in August 2019, up 19 percentage points from 61% in July 2018’. This is an important improvement as ‘Two-thirds of smart speaker owners value voice assistants’ ability to understand them above all other qualities.
Voice control in cars is reducing distraction for drivers; Systems such as ‘NVIDIA Drive’ are putting powerful voice control into cars to ‘enhance the driver’s situational awareness, assist in driving functions and provide intelligent interactions between the vehicle and its occupants’. This is now possible without an internet connection, which is often the case when driving.
Having digital assistants in our workplaces could make them more interactive and accessible. Google Home’s knack of recognising, and telling users apart, from their voices shows that coverage to multiple desks could be provided from a single device. Users could get a personalised service without needing to log in. Linking with office tools such as calendars, task lists, conferencing or messaging applications and room booking could make for powerful productivity improvements. I could ask this shared PA to ‘set up a project update meeting with my team this week’ and it would check calendars, find a suitable slot, create the meeting and invites, book the room and remind me before the meeting’.
Businesses seem to know about voice-activated digital assistants, with 85% of surveyed companies saying they’re aware, though few are actually acting on this. Of the same companies, only 14% are in trial or implementation, while 12% decided not to implement after a trial.
Privacy concerns around speech recognition, which people might brush aside in the home, are a much bigger hurdle for the workplace. Employees are understandably nervous about their employer being able to ‘listen in’ to what they say. Companies will also be concerned about the privacy of audio recordings made by voice assistants, which are mostly sent to third party cloud servers, because of the processing power needed to turn noisy clips of human speech into text. Given the confidentiality and data security requirements for many businesses, this method is likely to meet a big resistance from IT departments.
Although things are improving, communication with a computer is still far from a ‘natural’ conversation in most cases. From my experience with smart home systems, things work smoothly when I recite a known phrase. Saying ‘I’m leaving’ triggers a nice sequence of actions, turning everything off in the house, something I would previously have run around doing manually. However, a guest saying ‘I’m cold’ is likely to be met with ‘sorry I don’t know how to help with that’, rather than the system working out it should turn on the heating. You don’t need to hear that phrase too many times before you get frustrated and lose faith in the faceless voice. These issues will be addressed in time, but will limit the adoption of this technology for now.
Many Smart buildings still don’t satisfy the needs of users in a simple and effective way, or worse cause frustration. There are practical steps we can take to improve this, starting with integrating systems and using more sensors to provide more effective, flexible control, without needing users to input. Then thinking about the personal user experience; create fewer and simpler, personalised interfaces which ideally bring everything into one place, that’s familiar to users. Systems that learn from users can help to reduce friction and give people just the information and functions they need. Finally, voice control with natural language processing shows great promise for the future workplace, once privacy and data security have been addressed.
Communication between people and smart buildings will become more like communication between people, or better still, won’t be required at all.