Energy Management Analytics: What Are Reactive, Predictive & Prescriptive Analytics?
By Jack Bullock
Chief Engineer, PE, CEM, BESA
Modern systems, ranging from refrigerators to advanced HVAC systems, are designed to interact. Interaction comes in the form of smart technologies and connected sensors, providing a stream of information back to a given platform or system for use. Even old assets, like antiquated HVAC chillers or lighting systems, can be brought into the equation through the use of connected sensors and technology. The data collected in its raw form is almost meaningless. It shows energy use and runtime, but its real value becomes evident when combined to form big data and be processed through energy management analytics, notes Jeff Bertolucci of InformationWeek.com. School district leaders need to understand the triad of energy management analytics (reactive, predictive, prescriptive) and how they can be applied in educational institutions.
Reactive Analytics Identify Why an Event Occurred
The most common uses of analytics refer to how a situation occurred, known as descriptive or reactive analytics. Information gleaned from this type of analytics can be used to refine maintenance schedules, determine energy use behaviors’ impact on peak usage rates and more, reports Salvatore Salamone of Energy Central. Reactive analytics are a reaction to data, detailing when something occurred, the factors contributing to it, and its impact. This information holds great value for school leaders; it provides a baseline to understand the current state of energy use. Deployed in conjunction with information regarding student progress, reactive analytics further detail possible correlations between environmental conditions in the classroom and student progress.
Predictive Analytics Forecast Outcomes Based on Current Conditions
Understanding why something occurred is only a portion of the spectrum of energy management analytics. Predictive analytics take reactive analytics a step further.
As explained by Dian Schaffhauser of Campus Technology, predictive analytics allow school leaders to automate system controls and reduce energy expenditures by optimizing system settings. This is in consideration of internal and external factors, like energy use behaviors among students and variances in weather. The potential applications of predictive analytics include a better understanding of how current conditions impact asset runtime when conditions remain unchanged.
For example, predictive energy management analytics use data to determine the energy cost for running an HVAC unit throughout a winter storm. Similar applications predictive analytics may involve potential impact on energy costs and facility asset condition throughout emergencies and black swan weather events detailing what may happen for different scenarios. This allows for a limitless, "what-if" forecasting scenario to help education Facilities Managers understand the impact on the budget in the future and potential disruptions in the classroom setting.
Prescriptive Analytics Provide Direction to Achieve a Desired Result
After determining how current events will proceed, the third form of analytics defines the actions and interventions necessary to achieve a given outcome. This type of analytics is known as prescriptive analytics, and they prescribe a series of steps to reach the best outcome. In the case of energy management analytics, this would be cost savings through energy use reduction and increased energy efficient use behaviors.
Continuing the previous example, prescriptive analytics may define the settings necessary to improve the energy efficiency of an HVAC unit in place of other factors. School districts may leverage prescriptive analytics to accommodate maintenance needs including the prioritization of facility retrofits or the replacement of asset components based on runtime and asset function. This saves money by improving budgeting and lowering energy consumption costs simultaneously.