Data Science and Big Data Learning Algorithms and Analysis for Improved Operational Availability

Total equipment ownership costs and overall Fleet operational availability will be vastly improved by increasing equipment reliability, failure prediction, and overall operational readiness while greatly reducing the occurrence of unnecessary maintenance activities. The utilization of cost effective and open source Big Data advanced analysis software will allow the Navy to accurately predict equipment failure based on not only the equipments real-time diagnostics, but also trends of similar equipment behavior across the Fleet. Progeny believes integration of large and disparate formats of data, including historical diagnostic data and those obtained in real-time from in-situ sensing devices, into a normalized format will enable automated analysis by open-source big data learning techniques. Analysis of equipment degradation failure trends will allow our software to automatically recommend corrective hardware maintenance prior to failure occurring. Progenys proposed solution will leverage automated data acquisition, big data storage techniques, and automatic analysis of hardware life remaining to recommend corrective maintenance prior to a failure occurring. Our true innovation for this effort will be the creation of an automated CBM+ algorithmic capability which automatically receives component fatigue assessments, invokes algorithms to calculate the impacts to dependent hardware, and recommends condition-based maintenance.