Supply Chain Vulnerability Identification Using Big Data Techniques

NPS NRP Executive SummaryProject Summary: Marines operate the MV-22 globally with support from a multi-tiered maintenance and supply system. However, low aircraft readiness levels demand improved optimization of the current MV-22 sustainment system. With diverse datasets being generated across the MV-22 sustainment system, the data environment is ripe for the application of big data analytics (BDA) which has emerged as a discipline for deriving value from large heterogeneous data environments. This research specifically examined maintenance data for MV-22 aviation depot-level repairables (AVDLRs) that demonstrated high levels of reliability-related demand from 2016 to 2018. A cross-industry standard process for data mining (CRISPDM) was employed with the application of quantitative methods of analysis. First, outlier demand behavior for three AVDLRs was identified across aircraft, squadron, and serial number, with the use of Tableau. Second, popular, emerging, and anomalous failure themes contained within maintenance comments assigned the same malfunction code were differentiated using Lexical Link Analysis (LLA). Third, classification models were tested for predicting the intermediate-level (I-level) action taken codes (ATC) and program level failure modes, incorporating LLA comment classification categories as an additional independent variable. The findings of this research illustrate opportunities to derive deeper understanding of AVDLR failure and repair across multiple levels of naval aviation maintenance.Naval Research ProgramTopic Sponsor: N41LResearch POC Name: Eric Bach, CAPT, USNN4 - Fleet Readiness & LogisticsNPS-19-N120-AThis research is supported by funding from the Naval Postgraduate School, Naval Research Program (PE 0605853N/2098). https://nps.edu/nrpChief of Naval Operations (CNO)Approved for public release. Distribution is unlimited.