Nonparametric Recurrent Event Data Analysis

ABSTRACT: The Air Force has struggled for years to effectively evaluate the reliability of its fielded repairable assemblies and systems. This SBIR Topic sets aside the metrics and analytical methods currently used to make reliability dependent forecasts of future demands and repair parts. It seeks to find more appropriate tools and techniques that provide a capability for analysis of usage and event data to identify trends, determine root cause, and the best course of action to restore the reliability of degraded assemblies. When mature, the software package envisioned will provide a full suite of performance/repair data and new analysis tools with access for engineers and supply chain professionals. The data can also be algorithmically processed to provide decision support advisories to maintainers and supply chain managers. This capability will provide innovative business intelligence that will enable process improvements for budgeting, demand planning, failure forecasting, repair parts stockage, repair workload planning and estimating system time to failure. BENEFIT: The solution to the problem this SBIR Topic addresses has applicability throughout the Defense Department and Commercial Airline industries. Improved tools and techniques to evaluate the reliability of fielded systems for failure forecasting and projecting future requirements is the"Holy Grail"for industry worldwide. This subset for fielded repairable systems and their components is more complex and therefore has had less attention until recently. The benefits of this capability to the DoD are significant in terms of the potential to reduce the O & M budget.