A Data Quality Dashboard for Reliability Data

Product manufacturers and equipment maintenance organizations alike desire to understand the typical failure behavior of their machinery. One common approach is to perform a RAMS (Reliability, Availability, Maintainability, and Safety) analysis. A core element of RAMS is the statistical analysis of equipment failure data. While there are many established methods based on the parameter estimation of probability distribution functions, little thought is given today on the impact of data quality issues on those estimations. This is especially problematic as such issues are quite commonplace in industrial data. In this paper, we propose a data quality dashboard which identifies data quality problems and gives concrete advice on countermeasures. The dashboard design is motivated with an explanation of typical data issues related to reliability data based on five case studies as well as a review of the status quo of data quality assessment. We use data from the case studies to illustrate the benefit of our dashboard.