A capability approach for seismic risk analysis and management - Seismic Risk Analysis and Management of Civil Infrastructure Systems - page 258

for shelter insecure. Thus, an indicator that provided information about
whether a residence was permanent could be more accurate. The process
of normalizing indicators, which we discuss below, provides a way of com-
bining information from different dimensions of well-being into a compos-
ite account.
Scaling and combining indicators
Each indicator provides information about a distinct dimension of well-
being. To be able to combine diverse kinds of information together to form
a composite picture of the overall well-being of individuals, the process of
normalization (or scaling) is needed (Gardoni and Murphy 2009). In this
process, dimensionless indices called indicator indices (II) that range from
0 to 1 are created for each indicator. These IIs are based on a comparison
of the value of an indicator against reasonable maximum and minimum
values for that indicator. A value of 0 indicates that a hazard will have no
impact on a given capability. A value of 1 signifi es that a hazard will have
the maximum foreseeable impact. Finally, the IIs for diverse dimensions for
a given individual are combined with a procedure described in Gardoni and
Murphy (2009) and Murphy and Gardoni (2010), providing a composite
picture of the impact a hazard may have on well-being creating a hazard
index (HI) that also ranges between 0 and 1.
Putting the HI in context: hazard impact index (HII)
For a given value of HI, single individuals in a larger society will be impacted
less on average than single individuals in a smaller society (Gardoni and
Murphy 2009). This is because more people would share the impact and
there would likely be more collective resources for the recovery. For
example, what an HI means for China is different from that for El Salvador.
Therefore, the societal impact captured by HI needs to be converted into
an impact per capita by dividing HI by the size of the population affected
by the hazard, creating the hazard impact index, HII. Additional details on
the procedure for creating the HII can be found in Gardoni and Murphy
(2009).
Finally, as noted in the earlier section of this chapter, information about
the distribution of risks across a population is important for policy and
decision makers. To ascertain the likely distribution of adverse conse-
quences, a process of disaggregation can be used similar to that used by the
United Nations in assessment the level of development of sub-groups
within a community (Jahan 2003). In disaggregation, the HII for each sub-
group is computed and can then be compared against standards for accept-
able risk.