Champaign, IL — The National Center for Supercomputing Applications (NCSA) at the University of Illinois at Urbana-Champaign presented Caterpillar Inc. with the 1999 Industrial Grand Challenge Award on Wednesday, April 28. The award was presented during a lunch meeting of NCSA Private Sector Program Partners at the Radisson Suite Hotel in Champaign.
In keeping with NCSA’s mission to help improve the competitiveness of American industry, the Industrial Grand Challenge Award was established in 1992 to recognize corporations that accomplish competitive breakthrough applications as a result of their partnerships with NCSA. The 1999 award was presented to Caterpillar in recognition of the company’s use of Knowledge Discovery in Database (KDD) techniques to identify new truck engine customers and improve the quality and reliability of truck engines. KDD is the process of identifying hidden patterns and relationships within data. By employing data mining algorithms that can learn the buying behavior of current and potential customers, Caterpillar was able to not only classify those customers who were likely to purchase truck engines but also how many they were likely to buy.
In one project, SMART Selling, Caterpillar and NCSA researchers developed a process that merges data from the company’s own customer databases with public truck engine owner databases to identify high potential sales targets. Initial results from applying KDD techniques show an increase of more than 50 percent in the number of sales calls on potential customers. The average number of engines sold to truck fleet customers rose 67 percent. In 1998 the number of promising sales targets identified jumped to 35 percent, and the number of engines sold grew to 6.75 per truck fleet customer.
“We’re making more sales to more customers who are buying more Caterpillar truck engines,” said Doug Oberhelman, vice president responsible for the Engine Products Division. The engine marketing group used the customer knowledge obtained from using these techniques to develop two targeted direct marketing programs that generated a 17 percent average response rate–more than five times the rate that’s considered successful.
In a second project, Early Detection, Caterpillar and NCSA researchers used KDD to transform thousands of pieces of data describing product problems into easy-to-read visual maps and charts that clustered similar problems and indicated their relative significance. Now the company can analyze data and implement necessary fixes in a matter of weeks rather than months.
“By being able to more quickly and thoroughly analyze this data, we can resolve product problems more quickly, helping to protect and enhance Caterpillar’s reputation for delivering reliable, durable, quality products to customers,” said Dan Murphy, vice president responsible for the Performance Engine Products Division.
Larry Smarr, director of NCSA and the National Computational Science Alliance, said the award is well-deserved recognition for a long-time NCSA Private Sector Program partner.
“NCSA and Caterpillar have had a successful collaborative relationship for 10 years,” said Smarr. “The successes that have resulted from this relationship illustrate how our Private Sector Program is a tool for American business to use in maintaining and improving their competitive edge.”
This is Caterpillar’s second Industrial Grand Challenge Award. The company won its first award in 1993 for its use of virtual reality in designing earthmoving equipment.