ASA Journal Session: Technometrics
Session Slot: 2:00- 3:50 Tuesday
Estimated Audience Size: 100-150
AudioVisual Request: xxx
Session Title: Analysis of Events Defined by Duration and Severity
Theme Session: No
Applied Session: No
Session Organizer: Morris, Max D. Oak Ridge National Laboratory
Address: Bldg 6012, M.S. 6367 PO Box 2008 Oak Ridge, TN 37831-6001
Phone: 615-574-3140
Fax:
Email: morris@msr.epm.ornl.gov
Session Timing: 110 minutes total (Sorry about format):
110 minutes total....however you want Opening Remarks by Chair - 5 or 0 minutes First Speaker - 35 minutes or more Discussant - 15 minutes Discussant - 15 minutes Discussant - 15 minutes Floor Discussion - 10 minutes (or 5 or 15)
Session Chair: Morris, Max D. Oak Ridge National Laboratory
Address: Bldg 6012, M.S. 6367 PO Box 2008 Oak Ridge, TN 37831-6001
Phone: 615-574-3140
Fax:
Email: morris@msr.epm.ornl.gov
1. Events Defined by Duration and Severity, With an Application to Network Reliability
Lambert, Diane, Bell Labs, Lucent Technologies
Address: Rm. 2C-256, 600 Mountain Ave, Murray Hill NJ 07974
Phone: 908 582 6509
Fax: 908 582 3340
Email: dl@research.bell-labs.com
Becker, Richard, A., AT&T Laboratories
Clark, Linda A., Bell Labs, Lucent Technologies
Abstract: Communications networks are highly reliable and almost never experience widespread failures. But, from time-to-time performance degrades and the probability that a call from one city to another city (say) is blocked or fails to reach its destination jumps from nearly zero to an unacceptable level. High, but variable, blocking between the two cities may then persist for a noticeable period of time. Extended periods of high blocking or events can be caused by congestion in response to natural disasters, fiber cuts, equipment failures, and software errors, for example. Because the consequences of an event depend on the level of blocking and its persistence, lists of events at specified blocking and duration thresholds, such as 50% for 30 minutes or 90% for 15 minutes, are often maintained. Reliability parameters at specified blocking and duration thresholds, such as the mean number of events per year and mean time spent in events, are estimated from the lists of reported events and used to compare network service providers, transmission facilities, or brands of equipment, for example.This paper shows how data obtained with two-stage sampling can be used to estimate blocking probabilities as a function of time. The estimated blocking probabilities are then used to detect events and to estimate reliability parameters at specified blocking and duration thresholds. Our estimators are model-free, except for one step in a sampling bias correction, and practical even if there are hundreds of millions of observations. Pointwise confidence intervals for reliability parameters as a function of blocking and duration thresholds are built using a kind of "partial bootstrapping" that is suitable for very large sets of data. The performance of the algorithm for event detection and the estimators of reliability parameters are explored with simulated data. An application to comparison of two network service providers is given in this paper, and possible adaptations for other monitoring problems are sketched.
Discussant: Robinson, Jeffrey A. General Motors R&D Center
Address: 30500 Mound Rd 1-6, Box 9055, Warren, MI 48090-9055
Phone: 810 986 1101
Fax: 810 986 0124
Email: jrobinso@rmastat.ma.gmr.com
Discussant: Schmee, Josef Union College
Address: 1156 Rugby Rd., Schenectady, NY 12308-2423
Phone: 518 370 6248
Fax:
Email: Jschmee@worldnet.att.net
Address:
Phone:
Fax:
Email:
List of speakers who are nonmembers: None