MODELING TELECOMMUNICATION OUTAGES DUE TO POWER LOSS

Authors

  • Andrew P Snow IThe J. Warren McClure School of Communication Systems Management, Ohio University
  • Gary Weckman 2Industrial and Systems Engineering, Ohio University
  • Kavitha Chayanam IThe J. Warren McClure School of Communication Systems Management, Ohio University

DOI:

https://doi.org/10.23055/ijietap.2006.13.1.422

Keywords:

Telecommunications, reliability, point process, piecewise linear model, non-homogenous Poisson process

Abstract

This work involves analyzing reported US telecommunications outages due to power loss over an eight year period (1996-

2003). Data from the reports includes the dateltime of each outage, allowing time series reliability analysis. The arrival

process is modeled by two non-homogenous Poisson processes (NHPP): the popular power law process and a piecewise

linear process. The study concludes that the number of telecommunication power outages after September II, 2001 (9-11)

decreased. In this case, the frequency of outages changed abruptly, strongly inferring a change or breakpoint in fundamental

process. The major fmding is that there is strong evidence of reliability growth. However, the growth could not be

attributed to a monotonic "learning curve" improvement process. Rather, modeling provides strong evidence that 9-11 was

episodic enough of an event to act as the catalyst for improvement

Downloads

Published

2022-02-24

How to Cite

Snow, A. P., Weckman, G., & Chayanam, K. (2022). MODELING TELECOMMUNICATION OUTAGES DUE TO POWER LOSS. International Journal of Industrial Engineering: Theory, Applications and Practice, 13(1), 51–60. https://doi.org/10.23055/ijietap.2006.13.1.422

Issue

Section

Modelling and Simulation