The data science behind facility management
The role of today’s facility managers is expanding, involving them in more aspects of the business and making them more pivotal to a company’s success.
Many facility managers are now reporting straight to senior management, making FM data collection and the ability to understand and report on the data more integral to their role.
Big data, big opportunities
With the explosion of the IoT, it has become much easier to capture, transport and process granular “big” data from the workplace environment. Facility managers therefore no longer need to rely on traditional yearly surveys or consolidated reporting where the critical data gets lost – putting them at risk of making false generalizations or “drowning by averages”.
The rise of smart building technology alllows FM leaders to move beyond “gut feelings” and gain insights based on actual utilization data. This can be captured in real time by deploying wireless sensor networks and connecting to existing, and previously closed systems such as a BMS or energy meter infrastructure. Gathered in an analytics platform, the data can reveal new insights about employee preferences, facilities usage, and performance drivers.
This is where the new generation of analysts comes into play. These are not just facility experts, but data scientists who know how to interpret and visualize data in order to draw the right conclusions. For example, occupancy has been measured at peak times for decades but today’s more granular data allows to look for patterns and discover interdependencies by connecting with data from other sources such as complaints handling and helpdesk tickets, or the monitoring of temperature, humidity and noise levels. Mapping these correlations will help you understand why certain spaces are underused (e.g. poor air quality, too hot, too noisy,…) and take appropriate action.
Usage-based FM service delivery
Another IoT business case is cleaning, a task which has been around since forever, but has not been revolutionized much. Cleaning schedules are typically based on fixed frequencies, estimated on the basis of hi-level reporting. With sensor technology and mobile devices, it is now much easier to give real-time instructions to cleaners, driven by measured activity, instant feedback or cleaning-on-demand requests from building users. This way, cleaning becomes much more dynamic, and is targeted where it has the biggest impact on customer satisfaction. Service providers get more leeway to model the daily workflows with their customers‘ interests in mind. We call this new service model outcome-based cleaning.
Is it really as simple as that? For the users, yes. For the data scientists, no. A viable model also has to take into account the complexity of the work and contractual and other limitations. It needs to balance skills, tools, workloads, staff shortages, external factors such as weather conditions etc. with desired outputs and user satisfaction. Data science can help build realistic models and automate them. It essentially creates meaningful information from complex layers of data, making things easy and fluid for the workers on the floor, the building users and the facility professionals.
At MCS we work collaboratively with businesses to guide them from sensor to value. We provide end-to-end smart building solutions, combined with advisory services and data interpretation by an integrated team of data scientists and domain specialists. If you would like to know more about the possibilities, please don’t hesitate to reach out to us for a demo. We would love to hear from you.