Geoscience Reference
In-Depth Information
implications of the damage. Specifically for bridges, DM describes the damage to their
structural elements or the structural system. The final step in the assessment is to cal-
culate the decision variables ( DV ) in terms of mean annual probabilities of exceedance,
ν
( DV ). In general, the DV s relate to one of the three decision metrics that include, direct
dollar losses, downtime (or restoration time), and casualties. The DV s are determined by
integrating the conditional probabilities of DV given DM , p [ DV=dv
|
DM ], with the mean
annual DM probability of exceedance,
ν
[ DM ].
2.2. TRANSPORTATION NETWORK RISKASSESSMENT
The PEER methodology is limited to estimating the risk to components of the system.
In this section we present the network risk estimation method. Damage to the compo-
nents of the network often results in the closure of specific links until these components
are repaired. This action increases the level of congestion and travel time or reduces the
number of trips taken. Trip reduction is very difficult to predict, however, logical esti-
mates can be made given thesocio-economic profile of the area of study.
The increase in travel time can be found with respect to a baseline scenario. Travel time
delay, however, is highly correlated to the number of trips which are lost. The problem
of the risk assessment of a transportation network becomes more complex under this
approach, since the indirect loss has two components, the cost of the delays and the cost
of the losttrips.
In order to develop a network risk assessment model, it is necessary first to formulate
a traffic assignment model .A traffic assignment model allocates the traffic within the
components of the network based on the supply and the demand for trips. The results of
such an analysis are the flow and the timeneeded totravel through each component.
There are several traffic assignment models, however, the most popular models are the
fixed and the variable demand assignment. The formulation for these two models was
developed by Moore and Fan (2003) and is summarized in the Highway Demonstration
Project(Kiremidjianetal.,2006).Thereaderisreferredtothesereportsforfurtherdetail.
We briefly describe the twomodels.
The fixed demand model for the traffic assignment assumes that the demand between
eachoriginandeachdestinationisconstantanddoesnotchangeafterearthquakeevents.
The advantage of this model is that it is simple to use. The disadvantage is that it fails
when the demand greatly exceeds the capacity of the network due to its assumption of
fixed demand.
The variable demand model for the traffic assignment assumes that the trip rates are
influenced by the level of service of the network. If traveling becomes too expensive in
terms of time or distance, the users are expected to change their habits in order to avoid
the discomfort. After a major earthquake travelers will either have to accept the new
congestion levels or decide not to travel. It is difficult to estimate how many passengers
Search WWH ::




Custom Search