#

Using AI to learn your buildings’ energy behavior

  • Posted by admin on June 27, 2018

Buildings are a central part of the world today. We live, work and play in them. We socialize, learn and engage in them. In fact, it’s estimated that we spend 93% of our time in buildings.  Of course, buildings don’t remain static once the builder hands over the keys – their uses, occupants and components vary and change with time. And over time, we’ve also developed ways to manage these buildings and the operations within, including energy behavior and usage.  But the demands on facilities management and the cost of services have each continued to increase over the years and the industry demands relief.

Today, we are excited to announce the launch of IBM IoT Building Insights. This solution allows facilities managers to put the power of augmented intelligence (AI) to work in order to understand buildings and their energy behavior.

Data is increasing in size and complexity

The need for a solution like this stems from multiple challenges, the first of which is data overload. Just as the  complexities of facilities management has increased, so too has the volume of data being produced. The systems that allow modern buildings to function continues to expand. These systems manage energy, security, elevators, HVAC and lighting, all generating data primarily for the purposes of running each. At IBM, we have long recognized that this kind of data can be valuable for different reasons. Modeling, correlating and analyzing that data can generate transformational insights for building managers and tenants.  Companies that understand how their connected building operate will have huge competitive advantages in cost, productivity, and safety.

Energy continues to be a significant cost driver

One of the highest cost items for building managers is energy, with 26% of the cost of running a building dedicated to utilities (see Figure 1). Not only that, but 39% of all energy spend in the United States is consumed by buildings. It’s been clear for some time that addressing energy overspend is a clear area of savings for buildings.

Figure 1. Breakdown of costs associated with operating a building. Source: BOMA 2016 Office Experience Exchange Report (Office EER)

Introducing IBM IoT Building Insights

IBM IoT Building Insights, available on the IBM Cloud, focuses initially on energy optimization. It models buildings by connecting to meters and sub-meters, ingesting historical readings and then beginning to learn immediately. The offering presents a single view of the current energy profile across each building and the building subsystem, warning where there are unusual patterns of consumption.

IBM IoT Building Insights, through its responsive web app and hybrid mobile apps, will allow building managers to easily navigate energy usage across all buildings, and then zoom in to diagnose when buildings are consuming too much (or too little) energy, down to a subsystem level. Building managers can rank buildings in order of performance, and prioritize issues for engineers to address. Initial feedback from early trial customers has indicated that as much as 20% of costs can be avoided through a proactive deployment of the solution.

The first application of its kind

IoT Building Insights is the first application of its kind to leverage Brick — a uniform metadata schema for buildings. Using Brick, IoT Building Insights creates a knowledge graph of each building in an enterprise. It can quickly access this graph, along with AI analytics and sensor data, to create a baseline for each building. This can help to detect any anomalous behavior. This unique capability drastically decreases need for manual entry and custom adjustments to be made in the solution when systems or equipment within a building are changed or replaced. Instead it makes IoT Building Insights aware and responsive to the dynamic nature of every building in an enterprise.

Constantly learning about energy behavior

The solution begins learning with historical data ingested at launch, and then continues to learn every single day. The machine learning tools designed around the buildings model commits learnings and anomalies to memory. The system then continues to increase its intelligence as it learns more about the buildings, energy behavior, and surrounding context.

Central to the solution are the weather models native from The Weather Company (an IBM Business). With highly granular actuals and 48 hour weather forecasts informing all analytical models, the solution promises best-in-class accuracy for weather sensitive analytics. With energy consumption highly sensitive and correlative to weather patterns, as you’ve likely noticed in your monthly energy bills, this feature helps predict unusual energy patterns so team can then take action to pre-empt the spike in energy usage.

Learn more about IBM IoT Building Insights

To learn more about this innovative solution, download this two-page brief to further understand the benefits and how it works.

To talk to a sales rep or schedule a consultation, visit the IBM Marketplace page for IBM IoT Building Insights

Download a copy of the new IBM Institute for Business Value ‘Building intelligence into buildings: integrating artificial intelligence into building ecosystems’ report.

The post Using AI to learn your buildings’ energy behavior appeared first on Internet of Things blog.

Internet of Things blog

Leave a Reply

Your email address will not be published. Required fields are marked *