Civil Engineering Reference
In-Depth Information
Mid-term (weeks, months): the infrastructure progressively returns to a
new, different state of normal functionality (
recovery
).
Long term (months, years): the infrastructure is possibly upgraded/ret-
rofi tted with available resources to mitigate the risk from the next event
(
reconstruction
).
As far as the analyst's position on the time axis relative to the time-frame
is concerned, again three cases are considered.
Before the time-frame: the goal is
forecasting
the
impact
in order to
prepare mitigation measures.
Within the time-frame: the goal is
providing
managers
with
a
real-time
decision
support
system
(DSS), which updates the infrastructure state
based on the continuously incoming fl ow of information.
After the time-frame: the goal is
validating
models
against past events.
It is apparent how the goal of the analysis is strongly related to the analyst
position. It can also be observed how the second case is certainly more
challenging with respect to the fi rst and third, where the information basis
can be considered as stationary.
The third and fi nal time-related aspect has to do with the dynamic nature
of the infrastructure, i.e. with the fact that the impact of a disturbance
evolves with the elapsed time since the occurrence of a disturbance. Hence,
with respect to time, one important choice is whether the model is a
dynamic
or a
static
one. In a
static
model things are not really static but, rather, they
are stationary, meaning for instance that fl ow equations are formulated for
stationary conditions. This modelling choice of course has the potential to
impact the way different components within each system, as well as differ-
ent systems, interact, and, ultimately, how damage consequences propagate
within the infrastructure. Intermediate strategies are possible. For instance,
systems can be basically modelled as static (stationary fl ows), but consider-
ation of the different
characteristic
time
of each system is used to remove
the assumption of instantaneous damage propagation. A similar approach
is explored by Hernandez-Fajardo and Dueñas-Osorio (2011).
The spatial dimension of the problem is, as already mentioned, a further
important factor in the analysis, which is not independent of the described
time-related factors. Indeed, the spatial dimension evolves with time. For
instance, the spatial extent of interest is initially limited mainly to the local
strongly affected area (short term), and is then expanded to include later
an increasingly wide area covering adjacent regions, up to the national or
global scale (in the economic recovery phase) to contract again to the
national/local scales (in the reconstruction phase, and when long-term risk
mitigation actions are undertaken). In some cases, however, this widening
of scope can happen rapidly due to the propagation of the effects through
Search WWH ::
Custom Search