The U.S. National Highway Traffic Safety Administration (NHTSA)—part of the Department of Transportation—made a simple request of the Volpe Center and Transportation Information Project Support (TRIPS) contractors, led by CSC: To provide an efficient way to deliver car crash information in an open, public format following the U.S. federal standards and compliance rules.
The information, collected since 1997 under the Electronic Car Crash Data System, averages out data involving approximately 5,000 accidents every year. Data is collected in 24 sites around the country and consolidated in a large Oracle database (500 GigaBytes). NHTSA wanted to make the information available to several groups of people, including Department of Transportation researchers, insurance company investigators, vehicle manufacturers and the general public.
Previously, all data distribution had taken place using flat files and various proprietary formats; new federal regulations and technology evolution empowered the Volpe Center to move to the current open standard: XML.
Initially, the engineers at the Volpe Center explored using a conventional approach—SQL queries driven by Java code. They determined that moving data from a database into XML using SQL and Java would require about 50 Java classes and 150 SQL statements for one study type. Obviously, a project of this magnitude would be complex to design and debug and would require considerable maintenance of code.