PARIS --- Recent improvements in F-35 reliability figures are due to changes in the way failures are counted and processed, but do not reflect any actual improvement, according to the latest report by the Pentagon’s Director Operational Test & Evaluation.
The report notes several instances where the procedures for gathering and processing reliability data used by the program were needlessly changed to reflect improved results that do not exist in reality.
Three different types of data “massaging” are identified in the report: moving failures from one category to another, less important one; ignoring repetitive failures, thus inflating numbers of failure-free hours; and improper scoring of reliability. In all these instances, data reporting and processing rules were changed during the year for no other reason than to paint a more favorable picture.
The first example quoted in the report shows that changing failures from one category to another has artificially boosted reliability rates.
In June 2013, the Joint Program Office changed the way it accounted for “nut plate failures, one of the most common failures in the aircraft, as ‘induced failures’ rather than ‘inherent failures’” thereby excluding them from the calculation of MFHBF_DC.
Mean Flight Hours Between Failure, Design Controllable (MFHBF_DC) is one measure of reliability, and includes “failures of components due to design flaws under the purview of the contractor, such as the inability to withstand loads encountered in normal operation.”
This is how the OT&E report describes the issue (page 62):
“Nut plates are bonded to an aircraft structure and receive bolt-type fasteners to hold removable surface panels in place. One way nut plates can fail, for example, is when torquing a bolt down while replacing a removed panel, the nut plate dis-bonds from the aircraft structure, preventing securing the surface panel.
“Distinguishing between inherent design failures and induced failures can be subjective in certain cases. For example, if a maintainer working on the aircraft bumps a good component with a tool and breaks it while working on a different part nearby, it is a judgment call whether that is an inherent design failure because the component could not withstand “normal” wear and tear in operational service, or if it’s an induced failure because the maintainer was “too rough.”
The OT&E Office found that, between September 2012 and April 2014, records show “a generally increasing number of failures categorized as induced each month over the entire period, but a generally decreasing number of failures categorized as inherent for each month since April 2013. The decreasing inherent failure count per month is notable, as during this period, the F-35A fleet size and total hours flown per month were increasing steadily.”
Counting hours but not failures
The reliability of other parts was also evaluated using questionable methods.
“As of September 2014, an improved horizontal tail actuator component had been introduced and installed on roughly 30 aircraft out of a fleet of nearly 100. Failures of the older component were not being counted in the metrics at all anymore, but flight hours from all 100 aircraft were counted.”
This calculation, notes the report, “could result in the reported reliability of that component being increased by up to a factor of three compared to reliability if all of the horizontal tail actuator failures were counted.”
And, “as multiple components are being upgraded simultaneously due to the unprecedented and highly concurrent nature of the F-35 program, the cumulative effect on the overall observed aircraft reliability of the increased estimate of reliability from all of these components may be significant.”
The OT&E report also found other uses of questionable metrics to boost F-35 reliability, notably in how improper scoring resulted in “higher than expected [Mean Times To Repair] values.
“Discrepancies for which maintainers have to attempt multiple solutions before finding a true fix are being re-scored as a single event, while in the past they were documented as multiple repair attempts, each with its own MTTR.
“The individual MTTRs for these attempted repairs are now rolled up into the single, re-scored event,” thereby artificially reducing the number of reported failures although there has been no reduction in the number of failures.
Stealth materials need freezer storage?
The OT&E report also draws attention to the difficulty of repairing the F-35’s “low observable” materials, in other words the coatings that are designed to make it nearly invisible to radar.
Several factors likely contribute to extensive maintenance times, especially long cure times for Low Observable repair materials. The Program Office is addressing this issue with new materials that can cure in 12 hours vice 48 for example, but some of these materials may require freezer storage, making re-supply and shelf life verification in the field or at an austere operating location more difficult.
Click here for the F-35 section of the OT&E report (34 PDF pages) hosted by Defense-Aerospace.com.