Beyond Lighting: How Should We Use IoT for Building Controls?

May 18, 2016
Enlighted Inc.

Much of IoT technology is being conceived and deployed on top of lighting in buildings, and we’re desperately trying to understand and predict the effects of the collision between tools and products from two very different worlds – consumer electronics and commercial and industrial building controls.

The fact that the lighting grid offers a very convenient infrastructure for attaching sensors is certainly helpful, but many lighting practitioners justifiably fear that eventually lighting systems will be organized around data gathering rather than around the basic service of lighting. This could create an even worse general quality of light in buildings and further threatening current lighting industry business models. On the other hand, this also presents unlimited opportunity for new connections between design disciplines and greatly improved buildings in general.

Despite an endless stream of fantasizing and pronouncements about the future of lighting and the convergence of dozens of different classes of technologies in the built environment, we fail to recognize the limits of our ability to forecast or deliberately create the future. We need to recognize that much of what is emerging is accidental, random, and often a result of convergence rather than relentless specialization.

Thomas Patterson, a prominent lighting designer and deep thinker about technology in the built environment, in a recent LinkedIn post pointed out several problems in the current state of affairs. For one, he points out that there are very few correlations between the various features in IoT enabled lighting. He also notes that many specialized “innovations” and value adds can be done better and more cheaply with existing technologies. Lastly, he believes that complexity on the device level alone could bring some lighting manufacturing operations to a standstill due to the sheer number of SKUs and their related interoperability and product life-cycle issues.

Another issue is that in the rush to install sensors and IoT on top of the lighting grid, few are actually studying the interactive effects of lighting itself on the behavioral and other data that the sensors are gathering. Does highly responsive lighting really increase productivity, aid health, and pump up sales? If so, what are the causal factors? What specifically is “highly responsive lighting” – tunable, dynamic, exquisitely sensored, all of the above? We’re just beginning to ask the right questions in this area, let alone come up with usable answers.

The history of electrification offers many useful lessons- perhaps most surprising is that back in the early 20th century, we were also swimming in futuristic fantasies and pronouncements, and while many of them eventually came true, many of them were way off base. The fact that we’re still not much better at predicting the future never seems to stop us from trying.

Strange as it may seem, long before Edison’s bulb, the first use of electricity in the home was for a burglar alarm, patented by one Agustus Russell Pope in Boston way back in 1853. But after the initial provisioning of electrical service to the home for lighting, an explosion of new labor saving appliances soon followed, along with a rapidly expanding industrial base to produce the appliances and the emergence of planned obsolescence and consumer culture to support and sustain economic demand. Electrification created what we call the modern era.

In Electrifying America, Social Meanings of A New Technology, David Nye writes: “The adoption of new technologies in the home was in part a matter of personal prestige and conspicuous consumption, in part an expression of the belief in scientific progress concretized in a new vacuum cleaner or electric refrigerator, and partly an affirmation of the new roles for women as home managers.

Yet once these new machines had arrived, their meanings were manifold and unpredictable…tools and technologies bought for a certain clearly defined use develop other meanings over time as part of lived experience.

In The Meaning of Things: Domestic Symbols and the Self, two social scientists found, ‘Objects affect what a person can do, either by expanding or restricting the scope of that person’s actions and thoughts. And because what a person does is largely what he or she is, objects have a determining effect on the development of the self.’”

In the workplace, according to Nye, electrification was historically largely driven by nontechnical factors, specifically the emergence of large corporations with big capital to invest; an emerging engineer-manager class, and a large, mobile workforce with a high turnover rate largely due to lower level manufacturing jobs becoming numbingly boring lever pulling and button punching functions.

This high turnover rate was a crucial problem for the U.S. economy as electrification and automation came to manufacturing in the early 20th century, and resulted in the creation of our current health care system, among many other things, as corporations developed “welfare” strategies aimed at retaining workers.

With IoT in the home, men and women today face roles that are increasingly complex and mystifying as domestic device managers. Most urban households are double income in order to afford living in the densifying cities where the jobs are, time for domestic labor is shrinking, and technology seems to offer an answer.

But how to deal with “smart” systems that are expensive, at best intermittently interoperative, with short product cycles, poor customer support, and generally create overwhelming complexity? If you spend your whole day programming at the office, why would you want to come home and program all your smart devices? Parents also worry greatly about the amount of time the entire family spends on electronic communication devices and IoT in general, and yearn for a “simpler” time when they could roast marshmallows over the fire by the organic vegetable garden.

And in the workplace, many factors are contributing to a significant change in the nature of work itself, most obviously automation. And just like a century ago, corporations are scrambling to improve worker health, welfare, and satisfaction, as the labor turnover rate climbs again and workers are increasingly mobile in new senses of the word. Just as the iPhone started as a consumer device and became ubiquitous in business (a dramatic reversal of the typical technology product development path) user friendly controls and IoT may evolve first in the home and then migrate to the commercial sector.

Could technology convergence in the form of optimal user interface (just like the iPhone of course) offer the answer? Many people really want absolutely everything in the environment – cars, buildings, appliances – controlled by smartphones, and this makes a lot of sense in many ways. We all know that there are huge impediments to this scenario: standards, interoperability, energy use, complexity, warranties, and simple tech overload to name a few. But we like to assume we will overcome all of these, and maybe we will.

For now I’ll stick with buildings alone and do a little fantasizing myself, on what the ideal building control scenario might look like. Actually, I don’t trust my limited powers of foresight enough to really visualize fully, so I’ll just list what I think any new “paradigm” of building controls should accomplish and what particular problems it might actually solve:

1. Limited Choices – much of the iPhone’s success was arguably due to a deliberate decision to limit choices and functionality. This “pruning” is vital to avoid an increasingly common phenomena of choice paralysis. Engineers’ including dozens of buttons on home media controllers is the prime example of exactly what NOT to do with anything meant to be used by real human beings. When we’re faced with too many choices, we usually give up and do nothing.

2. Self Learning and Adjusting Control systems must become intelligent enough to be self regulating and to “learn” behavior patterns of any group of building users in order to optimize both comfort and energy and water use.

3. Based on Observed Behavioral Data – Inputs to the system must focus on underlying behavioral causes of energy use and must measure them objectively, minimizing self reporting, which can give inaccurate data. Tracking where building users travel, how they interact, and how they use spaces, for instance, provides a rich data set with which to design and operate better buildings.

4. Measure Interactive Effects – Do people perceive a space as being hotter if the color temperature of the light is “warmer” (as in lower CCT)? What behavioral effects, positive and negative, are directly caused by optimizing the energy use for lighting? How does changing building layout affect productivity, health, and sales? Measuring single metrics alone is easier, that’s why most systems are limited to single metrics alone. But measuring how one metric impacts another can pay huge dividends.

5. Balance Group and Individual Comfort Levels – Most existing building systems today are designed around a single baseline metric, like a 72° air temperature setpoint, that assumes an average comfort level in spaces where no single person is average.

A rational engineering approach (take the average and apply it to everyone) fails completely when applied to real humans in real buildings. An increased granularity of individual control over HVAC or lighting has been widely demonstrated to improve perceived comfort by itself even if it’s never actually used! This particular behavioral fact is the kind of thing that would never have been uncovered by surveys or caught in design briefs – there are undoubtedly many more crucial behavioral insights like this waiting to be uncovered by intelligent sensor systems.

6. Adapt to Existing Buildings – The economics of building projects dictate that new buildings are much more expensive to build that retrofitting existing buildings. Even with advanced design and construction techniques, 3D printing, advanced energy systems and many other innovations, this will most likely be the case for the foreseeable future. Also, retrofitting buildings is in general more energy efficient that building new ones. Light, flexible, efficient systems for lighting, IoT, telecom, and energy distribution will necessarily evolve around retrofit strategies, even though this is far from optimal in many cases.

I feel that we often over reach our efficiency goals, and that we’d be better off making 90% of existing building stock 50% more efficient rather than 10% of it 90% more efficient.

7. Be Compatible with Analog Control – Operable windows are, in my opinion, one of the best building envelope strategies because they are simple, low-tech, and afford great individual control. But they fell out of favor at one point in recent history because HVAC engineers and architects like to avoid liability for building skin failures by positively pressurizing building envelopes so that encroaching moisture is forced out.

New highly complex digital control systems for building services can and should take analog control into account and not discard it completely. No one likes losing their phone, and you shouldn’t be locked out of our house or lose access to vital services if you do.

If we’re able to achieve some, not even most, of the functional goals on the list above, we’ll start creating much better buildings. What will this do to us? If you consider that we possess bodies, brains, nervous systems, and emotions that evolved optimally for survival on the African savannah hundreds of thousands, even millions of years ago, a completely “frictionless” environment, where all you have to do is think of turning on the lights and they gently “glow up,” might actually drive us nuts.

Because of how evolution works, it will take a really long time for the effects of what we’re building now to express in changed phenotypes, and we certainly can’t imagine technology standing still long enough for that to happen. Here I’m getting into philosophy territory, and will save the complicated dive into morals and ethics for the next blog.