Data mining mine safety data

The U.S. Mine Safety and Heath Administration (MSHA) developed a mine accident database from Part 50 of the Federal mine safety regulations. This database is a valuable resource for keeping track of the numbers, rates and severity of mine accidents in the United States. Epidemiologists and mine safety researchers have used it to perform many analyses, helping to guide research and bestpractices. Database technology has changed from when the database was originally designed in the early 1980s. Contemporary analytic databases, typically called data warehouses, focus on table structure to facilitate integrating many different data sources for the purpose of applying analysis and visualization tools to gain more in-depth knowledge. In early 2006, the U.S. National Institute for Occupational Health and Safety (NIOSH) funded an exploratory research project to evaluate the potential benefits of reengineering the MSHA Part 50 data into a data warehouse environment and to assess the potential of data mining to give new insight. This paper explores the background of the Part 50 database, gives a general background of data warehousing and data mining, and presents some of the interesting analyses that resulted from a modernized Part 50 data warehouse using data mining.