Geoscience Reference
In-Depth Information
by Francis et al. (
1983
), Brandeau and Chiu (
1989
), Eiselt and Laporte (
1995
),
Hamacher and Nickel (
1998
), ReVelle and Eiselt (
2005
) and ReVelle et al. (
2008
).
19.4
Linkages Between Location Science and GIS
Location science problems have been studied in various forms for hundreds of years.
On the other hand, GIS was not developed to solve location science problems as
such. Their primary purpose was to collect, store, manage, manipulate, display
and analyze spatial data. In fact, for a long period of time, the two fields seemed
to develop almost independently. However, as practical problems became more
sophisticated, it emerged that GIS offered excellent possibilities to handle the spatial
data needed to solve these problems. As a result, GIS were initially viewed as a
tool to provide data to location science models and to visually present their results.
Indeed, there are numerous applications where GIS and location science models
were combined in a
loosely coupled
way in the sense that spatial as well as attribute
data were extracted from the GIS to be used by an already defined location science
model. The model was then solved by commercial software or some special purpose
procedure and the results were imported back into the GIS for visual presentation.
In this setting, data requirements were determined primarily by the location science
model and the main task was to consider the data structures utilized by the GIS
and the location science component and to develop a procedure for exchanging data
and results files between them. Examples of these early approaches are reviewed by
Church (
2002
), Church and Murray (
2009
) and Murray (
2010
).
Following the continual development of GIS, it became evident that the links
between GIS and location science could progress far beyond the concept of loose
coupling described earlier. It can be argued that the two fields are beginning to
converge in a number of ways, some of which are analyzed below.
19.4.1
Suitability Analysis and Data Generation
In many practical applications of location science a pre-processing stage is neces-
sary in order to assess all the potential sites for one or more facilities and select
those that meet a given set of pre-determined prerequisites for further consideration
.
As noted by Sumathi et al. (
2008
), apart from determining the set of feasible sites,
GIS may also provide a digital data bank for long-term monitoring of these sites,
for managing the collection operation and analyzing the routes between different
elements of the system. In addition, different data layers, each weighted by a
different factor, may easily be combined in order to calculate a suitability score
for each possible location and only consider locations whose suitability score
exceeds a pre-specified threshold. Several GIS offer interfaces to help determine
appropriate weighting factors. For instance, IDRISI's Decision Wizard includes a
