a) Accessibility as quality of life component (factors)
b) Accessibility measures
- Description of five GIS-based methods for measuring access
- GIS exercises with steps-by-step introduction to each of the
five methods (based on ArcView, ArcGis, and ArcIMS)
c) Technical requirements for conducting the exercises
- Data
- Specification of skills needed
- Specification of technology
·GIS OVERVIEW. Link to
www.gis.com as
one of the sources for basic information on GIS.
Development of the Training Module
The training module consists of a set of HTML based exercises and
associated data that can be downloaded from the project website. The
exercises were developed for ESRI's ArcView GIS. Prior familiarity of
GIS, specifically Arcview is recommended. However, the exercises are
designed such that, users not proficient in GIS can follow the easy
step by step directions. However, prior knowledge of computing is a
pre-requisite.
The inherent shortcoming of a web based training module is that it
is not possible to cater to varying level of understanding or prior
knowledge on GIS. Hence, the module had to be developed assuming that
the user is a novice in GIS, thus enabling a variety of users to
explore the possibilities of analyzing accessibility using GIS.
Another component of the training module is a live Internet Map
Server (IMS) that allows the users to interact with the data and
preview the results, beforethey performs the
analysis. This makes the user aware of the output created by each type
of exercise besides providing information on the input data and its
attributes. IMS helps visualize potential scenarios. Technical
development of the IMS website involved certain complications. Issues
pertaining to compatibility of the web server (IIS 5.0) and the
version of the servlet engine (Tomcat) with the spatial server (ArcIMS).
Future development envisions migrating the module to ArcGIS and
further customization of the internet map server.
Accessibility Analysis
Extending on the previous work by the project PIs (Emily Talen’s
work in particular), the team identified the factors that were
relevant for the analysis of accessibility to urban infrastructure and
services. The four key elements of the framework are: 1) places of
origin, 2) places of destination, 3) mode of travel, and 4) travel
route characteristics (Table 4). This framework recognizes the fact
that travel is not a necessary condition for access to urban
infrastructure and services, as some of them are accessed on site
and/or delivered. Cultural, institutional, legal, social, economic,
political, and other factors are recognized to exert significant
influence on the nature of infrastructure and service delivery and
consumption. Access is defined as the quality of having interaction
with, or passage to, a particular good, service, or facility. It is
understood in terms of physical accessibility measured as distance,
time, and mode of travel; in terms of socio-economic accessibility
measured as affordability; and in terms of other cultural and
institutional determinants of accessibility.
Table 4: Framework for Accessibility Analysis
Training Module Examples
The
training module is developed around the measures of accessibility that
are compatible with standard GIS functionality of Environmental
Research Systems Institute's (ESRI) software products ArcView and
ArcGIS. Only a portion of the procedures can be demonstrated in ArcIMS
context. The five accessibility measures include: container analysis,
covering, minimum distance, travel cost, and gravity potential.
Exercises are developed for each except the last method that is deemed
more suitable for analysis of access to markets and goods based on
competition and consumer preferences, and is therefore considered
inadequate for analysis of provision of public goods in the context of
developing countries. The exercises are all based on examples from
Cape Town, as its dataset was the most complete (Figure 2).
Table 5: Approaches to Measuring Access
(Double click to enlarge table)
Container
Container
Analysis allows for identification of the number of facilities
contained within a given spatial unit (e.g., census tract, city,
country, etc.). In the conceptual example below this analysis helps
count the number of schools within each of the three villages (Figure
3).
A similar example is used for the GIS-based exercise developed to
demonstrate this method. The exercise uses a sample of data from Cape
Town to understand the distribution of schools relative to census
enumeration areas (EAs), and to compare this pattern with the
educational attainment data by EA (Figure 4)
Figure 4: Distribution of Schools and Educational Attainment in
Cape Town by Enumeration Area
Covering
method identifies the number of facilities within a given distance
from a point of origin. In the conceptual example below the
educational centers within three villages are assessed with respect to
their access to health facilities. The area of covering is defined by
the user based on the substantive knowledge about the expected level
of service. In the example below, a radius of one mile is applied to
search for health facilities within that distance (Figure 5).
The GIS-based exercise applies this method in
examining the adequacy of fire service. Assuming a 3-mile radius as
reasonable distance to
allow
for emergency response, the exercise helps enumerate how many fire
stations are located within the 3-mile radius of the population
centers. The population centers are represented with
centroids to allow for this analysis, which requires point
designations of objects of interests. The following graph illustrates
the outcome of this analysis. It shows the number of fire stations
within and outside the buffered population centers, along with the
number of people served by fire services or deprived of them (Figure
6).
Minimum Distance
Minimum
Distance is the distance between a point of origin and the nearest
facility, based on traversing a real network. The conceptual
example below shows a table and a graphic output generated from the
minimum distance procedure (Figure 7). The highlighted table column
displays the distance between each school and its nearest health
facility. The graph illustrates that the minimum distance based on
existing transportation network sometimes yields results different
from what may be expected based on casual observation. Namely, a
facility that appears closer in aerial distance, may be further away
in terms of network distance and/or time needed to reach the facility.
Figure 8 : Service Potential of Schools Relative
to Closest Population Centers
Travel Cost
Travel
Cost is expressed as the average distance between a point of
origin and all facilities in the area. Using t he example of
health care facilities, this approach would explore the access to all
facilities from a point (household, population center, central
transportation facility, or another facility) in terms of distance
and/or time, and calculate the average value (Figure 9). Preferences,
transportation options, or other factors can be used as weighting
factors. If this accessibility value is calculated for many points of
origin, those points can be used to interpolate an accessibility
surface for a given facility type or all types.
The exercise developed for the travel cost method
calculates the number of population centers (i.e., enumeration area
centroids) that are within 3 miles from each police station. The
distance is measured through the transportation network (Figure 10).
Figure 10: Overlay between Enumeration Areas and
Three Mile Service Area around Police Stations
Web Application Assessment
Keeping in mind that the development of exercises,
web contents, and web interfaces is still work in progress, we can
make several observations about the work accomplished so far:
Clearly, the readiness for use of local data
received from Cape Town partners was crucial for our progress. The
variables taken for analytical exercises coincide partially with the
UN indicators. They analytical potential is there,
however, to help visualize important information about inter-urban
conditions and access opportunities.
Socio-economic data crucial for understanding the
segregated nature of access has only been obtained recently. The
information on age and racial distribution will be incorporated in the
final versions of this training module. In addition to schools, fire
stations, and police stations, the exercises will be complemented with
information on other community facilities and utilities (access to
water and sewer, for example), and open space.
Reliance on only one site prevented our observation
of the complexities specific for other urban environments in
developing countries. The improvement in data contents from Katmandu
and Gaborone will compensate for this, and exercises with data from
those regions will also be included. The feedback
from the partners point to significant contextual differences in terms
of community problems, issues, and their manifestation.
While the GIS exercises and hard data will capture
some of those contextual differences, it seems that in order to
accurately depict the local conditions, it may be necessary to obtain
or develop conceptually diverse databases. For example, the
coincidence between the level of income and access to services that is
shown in Cape Town, would not hold true in Gaborone or Katmandu. The
access inequities are more of a micro scale and more peculiar in
environments that are not experiencing as high socio-economic
segregation.
The contextual nature and differences will pose new
methodological challenges and will certainly offer new learning.
Finally, going back to the original project purpose and objectives,
the development so far was clearly geared toward providing GIS
training. With our specific focus on accessibility, the general
purpose of the project is achieved. The question of who would ideally
be the users of this module is important to resolve. The project
partners already have GIS capabilities and skills that the module
exercises are trying to demonstrate. For example, reports from Cape
Town City Planning and Information Services units offer comprehensive
profiling of their communities. (The reports are referenced below.)
Similarly Katmandu Regional Valley Authority has issued a similarly
comprehensive GIS-based report. (Reference also provided below.)
The exercises focused on accessibility promise to be useful for the
employees in those government agencies, as well as the community
groups and organizations that are interested in taking advantage of
available technologies and information streams. The technology
transfer component was only indirectly addressed during the course of
this project. However, in the still forthcoming final phase, we hope
to gain additional insights about the utility and usability of the
training module.
GIS-based Reports Available (or Produced) by the Partners
Information Services, CMC Administration, Urban
Policy Unit, Cape Town Administration, and City of Cape Town. 1996.
THE PEOPLE OF THE CITY OF CAPE TOWN, A socioeconomic profile of the
metropolitan area. Summary data from 1996 census.
Urban Studies, Surveys and Land Information Branch, City
Planner's Department, City of Cape Town. 1996. LEVELS OF LIVING IN
THE CAPE METROPOLITAN AREA _ The social health and well-being of the
communities of the Cape Metropolitan Area.
Information Services CMC Administration. 1997. CAPE METROPOLITAN
AREA 1997. October household survey.
Previous page |
Next page
PROJECT
SUMMARY
