Elizabeth Byrnes and Jan Aikins
The Innovative Applications of Artificial Intelligence conferences highlight the most successful applications of AI techniques to real-world problems. The applications selected this year continue to illustrate the maturity of AI as a commercially viable set of technologies. AI has infiltrated a huge number of companies around the world and has enabled those companies to solve significant business problems and achieve major business benefits. The sixteen winners of the sixth annual Innovative Applications of Artificial Intelligence conference are a representative international sampling of the critical and growing contributions AI brings to business.
This year’s IAAI applications represent a cross-section of large-scale organizations and tasks. Industries represented include computers, finance, law, manufacturing, health-care, publishing, telecommunications, as well as government agencies. Functions that have been AI-automated range from administration, accounting and finance, to customer sales and service; from software re-engineering, to transaction processing. They cover product design, customer service, diagnosis, manufacturing, quality management, and production. As a whole, the applications show the continued movement of AI into critical core business functions.
The companies chosen to present at this year’s conference are: AT&T, Australian Department of Veterans’ Affairs, Bell Atlantic Corporation, Countrywide Funding, DHD Systems, Inc., E. I. duPont de Nemours & Company, Inc., EDS, EKATO, Fraunhofer Institute for Information and Data Processing, IBM, Inference Corporation, Information Technology Institute, IRS., Lockheed, Mendall Associates, Pacific Bell, Singapore Press Holdings, SoftLaw Corporation, SUNY at Stony Brook, Systems Development and Analysis, The University of Texas, US Navy, and the Vitro Corporation.
This year’s winning systems use not only rule-based reasoning techniques, but also techniques like case-based reasoning, natural language, text recognition, and model-based reasoning to accomplish their tasks. All winners were selected for their successful uses of the technology to solve real-world problems with demonstrable results.
DuPont’s winning application, for instance, which schedules overtime labor in a manufacturing unit, is saving $30,000 per week in overtime pay in one plant alone. In addition, worker’s grievances filed with the union for overtime reasons have plummeted. Another winner, the Australian Department of Veterans’ Affairs’ claims processing system, developed by SoftLaw Corporation, has reduced the claims handling staff from nine to one, and claims processing time from 180 days to 20.
But these achievements are not isolated. As you read through this book, you will discover other applications that have produced similar rewards. For example: Lockheed won with ASAP, an advisory system for automated procurement that has reduced the error rate from 75% to absolutely zero for the first 12,000 documents generated in the first two years of its use.
To help with the sales-service negotiation process, Inference and Bell Atlantic Corporation created SSNS, an application that provides automated support of needs-based selling. The application has increased sales efficiency, decreased training requirements, optimized customer contact time, and improved accuracy and efficiency.
From EKATO and IITB in Germany comes the COMIX system, an application that assists the sales department in designing a mixing machine that fulfills customer requirements. Singapore Press Holdings and Information Technology Institute has automated the layout of advertisements for a leading newspaper with ALEXIS, a system that helps the company in its move toward a virtual publishing environment.
Lockheed’s Clavier is a case-based reasoning system that assists in determining efficient loads of composite material parts to be cured in an autoclave. The system is being used daily on Lockheed’s shop floor and has virtually eliminated the production of low quality parts that must be scrapped, saving thousands of dollars each month. As one of the first fielded case-based reasoning systems, Clavier demonstrates CBR to be a practical technology that can be used successfully in domains where more traditional approaches are difficult to apply. AT&T and Inference teamed to create Trouble Shooter, a component of the Expert Solutions Platform that employs advanced technologies (including case-based reasoning) to assist in problem diagnosis and information retrieval at the National Service Assistance Center.
Pacific Bell has millions of telephone calls that cannot routinely be charged. To correctly bill such calls, investigators must manually search through telephone records. With the help of Inference Corporation, Pacific Bell now employs EMCS, an expert message correction system that is currently processing thousands of unidentified calls each day.