The Army says it is making headway with an innovative data-processing technique meant to automate maintenance and support of vehicle inventories and weapons systems.

The parametric data reduction tool, PaDRT, is a product of the U.S. Army Communications-Electronics Research, Development And Engineering Center (CERDEC) and Penn State University’s Applied Research Laboratory. After four years in beta, researchers recently moved it to the U.S. Armament Research Development and Engineering Center (ARDEC) at Picatinny Arsenal for pre-deployment testing.

“The issue is that we have too much data. We have had to find ways to reduce the data, but we don’t want the data on mission-critical systems to be compressed or reduced,” said Bill Ward, CERDEC’s project lead on Condition-Based Maintenance Plus (CBM+).

The CBM+ effort aims to build an end-to-end infrastructure to collect, store and transmit information on the health and usage of a range of assets, in order to drive efficient maintenance and aid in asset planning. Data reflects the current use of the systems and can potentially identify vulnerabilities and flaws before they arise.

Modern vehicles and weapons systems are equipped with a vast array of sensors to track a diversity of functions. CBM+ has suffered, not from a lack of input, but rather a deluge. With the emergence of PaDRT, planners say they will be better able to get their hands around this mass of information.

“Now we can look at that raw sensor data coming off the weapons system and strip away extraneous data. It can look at just one-tenth of the data, it can take a snapshot of a single moment when the system goes beyond the expected parameters,” Ward said.

The art here is in knowing what to get rid of and knowing what to keep. “You have to retain the original data set. You don’t want to use a data reduction tool where once you strip it away, it’s gone,” he said.

With that original data safely stored, the system will then start scanning for aberrations. “If it is operating within the operating profile, I don’t need to know that. Just show me one data point that says I am within tolerance,” Ward said. “But when there is an aberration, show me before, during and after in detail, so that I can analyze that.”

To pare away the bulk of redundant information, planners are looking to subtract the normal, and focus on what remains. “When you are reading a log and all of a sudden there’s a hiccup, that is the point that I want to see,” Ward said.

CBM+ has been a priority in recent years across the military services. Naval Sea Systems Command, for instance, uses it to streamline maintenance across the fleet, noting in a 2016 document that CBM+ “allows sailors time to focus on other duties while simultaneously sustaining the reliability of their equipment.”

In a policy document issues this spring, Air Force said it would turn to CBM+ “in the selection of maintenance concepts, technologies and processes for all new weapon systems, equipment and materiel programs based on readiness requirements, life-cycle cost goals, and … functional analysis.”

While CBM+ data management techniques have so far applied largely to vehicle and weapons systems, Ward said CERDEC is exploring ways to apply theme same algorithms to C4ISR systems.

“It would give us critical information on electronic systems, some actionable data about a communications device that is showing a likelihood of failure so you can get those replacement parts on hand before it fails,” he said.

There is a certain urgency in bringing more effective maintenance to C4ISR assets. “When comms go down you are pretty much dead in the water,” Ward said. “This is a mission-critical asset and the Army realizes that this is one of those elements that has to be CBM-enabled.”

To bring CBM+ to the C4ISR community, researchers will need to dig deeper still into the subtleties of data. Usage information on electronic systems will look somewhat different than the data on mechanical systems they are used to seeing. Ward suggested new data processing techniques likely will be needed to automate and enhance maintenance around communications and intelligence resources.