Facility leaders share their thoughts on what to expect this year and beyond
Join Dave Thompson on Feb. 27 in our Ask the Expert session on motivating and recognizing technicians and janitors
In today’s facility management environment, data is king. Data about maintenance schedules, operating targets, feedback from smart devices and building management systems, manufacturer documentation — the list goes on. But amidst all that data, the most critical information can be missed. Consequently, data’s value in helping facility managers drive down costs across a portfolio is rarely maximized. But with the right processes in place to ensure reliable, accurate facility data, more predictive, data-driven operation are possible, so that facility data can be used to drive down costs.
Historically, maintenance schedules have functioned as the primary source for such routine jobs as changing filters or checking fire extinguishers. In practice, this does provide a structured way to gather asset performance data and assure a standardized level of maintenance activities. However, this approach to maintenance is rarely optimized for either asset performance or cost control.
With more sophisticated methods of data collection readily available today, predictive maintenance can significantly improve both asset performance and control costs by providing real-time analytics on a specific asset’s status. For example, pressure gradients across filters can allow replacements to be timed to actual operating conditions. Monitors on fire extinguishers can identify which units are actually outside their required operating ranges. Temperature monitors on air handlers can indicate when dampers have failed to deploy correctly.
The key to moving to this more predictive, data-driven operation is twofold:
1. Making sure accurate data can be easily accessed and exists in an easily utilized format.
Much data delivered to the facility manager has typically been provided in an unstructured format. Such items as warranty documents, maintenance manuals, as-built drawings, and asset specifications are often in a printed format. Even when delivered electronically, they tend to be submitted as PDFs or other non-readable format. Compounding the problem is that these documents quickly become outdated or completely obsolete. Most have never been updated or converted to computer-readable format even as buildings get remodeled, furnishings moved, and equipment replaced. On the electronic building data front, the sheer volume of data points provided by building systems is overwhelming and thus mostly ignored.
Therefore, the key to these issues is having ready and filtered access to accurate data that the facility manager can trust. Both parts of the previous statement are necessary — access and trust. Even when facility managers install new computer systems and get access to all the available data, they often have no confidence that the data is correct. Remember the value of data is its accuracy and the key to having accurate data is not the technology itself but the processes and procedures that facility managers have in place for getting and maintaining data accuracy.
2. Making sure to have the right data.
Just as important as having access to accurate and accessible data is to know what data is needed. This is really a bigger problem in today’s connected world than the first item because data overkill overwhelms many managers precisely when it’s decision time. Needed information, for example, may be predictive information such as vibration and temperature monitoring to detect an upcoming problem, but in many facilities literally thousands if not millions of data points may be created each day.
Yet, the problem has a specific solution. The data issue usually seems overwhelming only if a management strategy has not been defined to focus, at least initially, on the most critical assets or at least those with the highest opportunity cost. The point is to identify data sets that provide increasingly better guidance as to an asset’s performance.
Reliable Facility Data Essential To Predictive Operations