Visualization of big data through ship maintenance metrics analysis for fleet maintenance and revitalization

Abstract

There are between 150 and 200 parameters for measuring the performance of ship maintenance processes in the U.S. Navy. Despite this level of detail, budgets and timelines for performing maintenance on the Navy's fleet appear to be problematic. Making sense of what these parameters mean in terms of the overall performance of ship maintenance processes is clearly a big data problem. The current process for presenting data on the more than 150 parameters measuring ship maintenance performance costs and processes, containing billions of data points, is still done by static, cumbersome spreadsheets. The central goal of this thesis is to provide a means to aggregate voluminous maintenance data in such a way that the causal factors contributing to cost and schedule overruns can be better understood by ship maintenance leadership. Big data visualization software was examined to determine if visualization tools could improve the understanding of U.S. Navy ship maintenance by its leaders. This thesis concludes that the visualization of big data supports decision making by enabling leaders to quickly identify trends, develop a better understanding of the problem space, establish defensible baselines for monitoring activities, perform forecasting, and evaluate metrics for use.