Reducing business surprises through proactive, real-time sensing and alert management

  • Authors:
  • Mitchell A. Cohen;Jakka Sairamesh;Mao Chen

  • Affiliations:
  • IBM T. J. Watson Research Center, Hawthorne, NY;IBM T. J. Watson Research Center, Hawthorne, NY;IBM T. J. Watson Research Center, Hawthorne, NY

  • Venue:
  • EESR '05 Proceedings of the 2005 workshop on End-to-end, sense-and-respond systems, applications and services
  • Year:
  • 2005

Quantified Score

Hi-index 0.00

Visualization

Abstract

OEMs need to transform the way they do business in order to ensure better quality of products and services. Crucial failure symptom information is lost between the end consumers of products and the manufacturers. Manufacturers have access to this information but are typically unable to handle its volume in a timely fashion. However, using this data properly can result in diminished labor time in issue resolution, decreased warranty costs for manufacturers and improved customer retention. In this paper, we present a novel system for Proactive Real-Time Event and Alert Processing, which enhances an enterprise's ability to monitor, analyze and detect critical business events and situations. This capability can help improve operational efficiencies, reduce costs, streamline processing of business alerts, and enable the enterprise to react in a more timely fashion. The system can enable monitoring of near real-time, low-level, industrial sensor and controller events, and high-level events from underlying structured and semi-structured data. The alert system uses domain knowledge to enable processing the events in real-time, performing the appropriate analytics and evaluation on the events and alerting the right set of users. The alert and event processing system has been deployed and validated in real pilots with industry specific data. We are currently validating the scalability and performance of the alert system with many different information sources and high volume sensor information.