RESEARCH
A New Approach to ITM
As a project to standardize inspection, testing, and maintenance data nears completion, experts look to a future of more effective, data-driven codes and standards
BY ANGELO VERZONI
A new, scalable approach to standardizing and consolidating inspection, testing, and maintenance (ITM) data may soon bring better-informed ITM frequency requirements to a range of NFPA documents—requirements based on data-driven, risk- and performance-based criteria rather than anecdotal evidence.
That’s the aim of the Fire Protection Research Foundation (FPRF), which is nearing completion of a project that could enable large-scale ITM data collection, the standardization of ITM requirements in codes and standards, and ease of sharing ITM data among industry stakeholders.
Nearly 70 NFPA codes and standards require some form of ITM for fire protection systems, and an important aspect of those requirements is knowing when to perform these activities. Even so, the ITM frequencies required by standards such as NFPA 25, Standard for the Inspection, Testing, and Maintenance of Water-Based Fire Protection Systems, aren’t based on empirical data.
“There isn’t a lot of substantiation to those requirements,” said Victoria Hutchison, a research project manager with the FPRF. “They’re based on engineering judgment and guidance from technical committee members and their experiences in the field.”
While there’s a great deal of proven value in that field-based experience, experts say the requirements could be made even more effective with the addition of data-driven approaches to ITM.
“There’s a huge need for this data,” Hutchison said. “We see this initiative not only as an opportunity to enable facility managers to make data-informed decisions on ITM activities for their own facilities, but also as an opportunity for stakeholders to provide a standardized set of data to do further research on. This research could help us do things like predict when a fire pump is going to fail.”
Standardizing the data
ITM data has always been collected by various entities, including contractors, inspectors, facility owners, insurers, and manufacturers, but rarely in a consistent way. It’s an issue the foundation has run into many times during past projects.
“We’ve tried to collect ITM data to analyze the performance and reliability of fire protection systems,” Hutchison said. “But one of the biggest challenges has always been that everyone you get data from has it in a different format, and the terminology is different. It’s an extremely manual and laborious process to consolidate all of it into a single format that can be analyzed.” The new project, officially named the ITM Data Exchange project, was launched to fix the problem, and after more than a year of research, the foundation has come up with what it sees as the ideal approach to structuring ITM data to best facilitate data exchange and analytics in the future.
Hutchison explained the approach in an article she wrote for Facility Executive magazine in December. “Have you ever sketched out an idea, problem, or process on a whiteboard?” she asked. “If you’re trying to explain the flow of information, a process, or key entities and the relationships between them, I bet you start drawing circles and arrows on your whiteboard to convey this information. The circles likely represent the people, places, or items you are talking about and the arrows represent the relationships between them.”
A fire pump, for example, could be represented by a circle, with arrows connecting that circle to other circles representing the pump’s pressure, rated flow, manufacturer, and so on. This concept is known as a graph data model, and it’s the way FPRF researchers see ITM data being structured in the future. The advantages to the model, Hutchison said, include being able to easily visualize how individual elements of data relate to one another and the flexibility of data input and analysis. “If someone wants to analyze a hundred data elements, they can do that,” she said. “If someone wants to analyze only five, they can do that, too.”
The model also allows users to easily look at ITM activities from different perspectives. They can look at data for one specific fire pump or for a specific building or for all fire pumps from one specific manufacturer or for a specific building over a specific date range.
“That’s the beauty of the model,” Bart van Leeuwen, founder of Netage BV, a knowledge graph company that worked with the foundation on the data exchange project, said in a video created in January. “We’ve seen from data that sometimes the test and the inspection were done in the same run,” and the data model allows you to see that “without doing really ugly, ugly things.”
Netage BV founder Bart van Leeuwen explains the ITM data exchange model (Netage BV)
“With stakeholder participation in future data collection efforts, our long-term aim is to provide recommendations to technical committees on what ITM frequencies should be,” Hutchison said. “These frequencies could address the breakdown of specific components, when they should be replaced, how often they should be inspected, whether pumps are running too much or not enough—a lot of questions we’re hoping we can provide analysis on. This data model brings us one step closer to being able to harness the true value and transformative power of ITM information.”
The results of the ITM Data Exchange project are expected to be released in May.
ANGELO VERZONI is a staff writer for NFPA Journal. Follow him on Twitter @angelo_verzoni.