How Microsoft quietly built the city of the future

Until recently, Microsoft was using disparate building management systems to manage 30,000 unconnected, sensor-enabled pieces of equipment. Imagine a symphony orchestra, but with every musician playing from different sheet music. Then, imagine trying to conduct that symphony – to make sure the music was on tempo, in key and starting and stopping as it should.

Smith’s team was on a journey to find harmony.

When Smith, Jay Pittenger (Smith’s boss) and others started exploring ways to manage buildings smartly, they realized it would cost upward of $60 million to “rip and replace” enough equipment to get those 30,000 sensors to whistle the same tune.

This would not only involve costly construction and equipment replacement, but it also would mean displacing employees and losing work while teams temporarily shut down labs. Smith and team knew there had to be a less pricey, less disruptive way to achieve data harmony, but after a whole lot of looking, they couldn’t find one.

So they invented one.

Smith’s team enlisted the help of three vendors in the field of commercial building data systems and created a pilot program in 13 of the buildings on Microsoft’s Redmond campus. The team developed an “analytical blanket” to lie on top of the diverse systems used to manage the buildings. The blanket of software finally enabled equipment and buildings to talk to each other and to provide a wealth of data to building managers.

“It hasn’t been a bowl of cherries – my hair wasn’t as gray before we started,” Smith says. “The challenge with building systems is that they can create a lot of chatter from multiple systems, but there’s value there if you connect and capture it. It’s all about the data. If you can’t get data out of the buildings, you’re done.”

The new tool did get data out of the buildings – great tidal waves of data that came cascading into the ROC, telling engineers about everything from wasteful lighting schedules to hugely inefficient (but up until then, silent and undetectable) battles being waged between air conditioners and heaters to keep temperatures pleasant.

In one building garage, exhaust fans had been mistakenly left on for a year (to the tune of $66,000 of wasted energy). Within moments of coming online, the smart buildings solution sniffed out this fault and the problem was corrected.

In another building, the software informed engineers about a pressurization issue in a chilled water system. The problem took less than five minutes to fix, resulting in $12,000 of savings each year.

Suddenly, the symphony of sensors was not only following the conductor, its musicians were all playing the same song. As buildings came online and data poured in, it created what engineers called a “target-rich environment” for problem solving. Smith and the team soon expanded the pilot to a handful of additional buildings, and by summer’s end they plan to have the whole Redmond campus online.

The team now collects 500 million data transactions every 24 hours, and the smart buildings software presents engineers with prioritized lists of misbehaving equipment. Algorithms can balance out the cost of a fix in terms of money and energy being wasted with other factors such as how much impact fixing it will have on employees who work in that building. Because of that kind of analysis, a lower-cost problem in a research lab with critical operations may rank higher priority-wise than a higher-cost fix that directly affects few. Almost half of the issues the system identifies can be corrected in under a minute, Smith says.

The change has created groundbreaking opportunities for Smith and his team.

“Our conversations have changed,” Smith says. “Before, the calls we got were about buildings being too hot or too cold, or about work orders. Now we’re talking about data points and building faults and energy usage. We’re seeing efficiencies that we never even contemplated when we started this journey.”

​Image courtesy of Microsoft.