From ceginfo@u.washington.edu Thu Aug 23 13:19:42 2001
Received: from jason04.u.washington.edu (jason04.u.washington.edu [140.142.8.53])
	by lists.u.washington.edu (8.11.2+UW01.01/8.11.2+UW01.04) with ESMTP id f7NKJd0118924
	for <cegrads@lists.u.washington.edu>; Thu, 23 Aug 2001 13:19:39 -0700
Received: from homer32.u.washington.edu (ceginfo@homer32.u.washington.edu [140.142.8.42])
	by jason04.u.washington.edu (8.11.2+UW01.01/8.11.2+UW01.04) with ESMTP id f7NKJcr02748;
	Thu, 23 Aug 2001 13:19:38 -0700
Received: from localhost (ceginfo@localhost)
	by homer32.u.washington.edu (8.11.2+UW01.01/8.11.2+UW01.04) with ESMTP id f7NKJaQ101610;
	Thu, 23 Aug 2001 13:19:37 -0700
Date: Thu, 23 Aug 2001 13:19:36 -0700 (PDT)
From: Civil and Environmental Engineering <ceginfo@u.washington.edu>
To: <faculty@ce.washington.edu>, <cegrads@u.washington.edu>
Subject: Re: MSCE Defense Jason Holdridge 8/28/2001 (fwd)
Message-ID: <Pine.A41.4.33.0108231318340.92776-100000@homer32.u.washington.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII

> The final examination for the MSCE degree for Jason Holdridge will be held
> on Tuesday, August 28, 2001 at 2:00 p.m. in More 119.
>
> Efficient Estimations of Crash Severity Models to Assess the In-Service
> Performance of Roadside Hardware
>
> Abstract:
>
> In the U.S., on average, single vehicle run-of-the-roadway crashes
> account for about one million highway crashes each year.  Although this
> accounts for about 14 percent of all reported crashes, these crashes
> result in about one third of all highway fatalities.  A great portion of
> run-off-the-roadway fatalities occurs from collisions with fixed objects
> on the roadside.  While roadside hardware is meant to prevent fatal
> injuries from collisions with more dangerous fixed objects such as
> utility poles, and trees, hardware itself can contribute to fatality and
> disabling injury risks.  Controlled crash-testing of roadside hardware
> has been in practice for over half a decade to design for effective
> mitigation of run-off-the-road crash frequencies and severities.  In
> spite of a significant body of knowledge compiled over half a decade,
> debate still occurs, especially at the state department of transportation
> level, as to whether crash testing-based knowledge is adequate as a
> standalone resource.  Sufficient argument through limited empirical
> analyses has been posited for the need to complement crash testing based
> knowledge with in-service performance-based knowledge of roadside
> hardware.  The need for a cogent argument is comprehensive, and the
> formulation of the argument should rely on sound mathematical analysis of
> data on in-service performance of roadside hardware.  This thesis
> attempts to be the first comprehensive effort to develop robust
> econometric formulations of in-service performance of roadside hardware
> in one important dimension, crash severity.
>
> Through development of efficient statistical models via the nested logit
> formulation, this study attempts to provides insights into the relative
> performance of roadside hardware.  This is achieved by controlling for
> geometric features, environmental and temporal characteristics, vehicle
> characteristics, and driver characteristics.  While the econometric
> formulations of in-service performance are the major contribution of this
> thesis, systemic issues relating to cost-effectiveness of roadside
> hardware retrofits are also examined.  As an example, an empirical
> cost-benefit analysis of bridge rail retrofits in Washington State is
> conducted.  By assessing different bridge rail lengths and their
> associated costs and benefits, we investigate the presence of
> cost-optimal bridge length thresholds for retrofitting.  The sum import
> of this thesis is the identification of pertinent information in the
> vehicle, roadway and human dimensions to allow roadway engineers to focus
> on key gaps between vehicle and roadside designs.
>
 Professor Venky Shankar, Advisor
>
>
>
>
>


.