|
|
GIS Toolkit
A
guide to planning, designing, implementing, and managing your GIS
The decision to invest in a GIS is a major one for
any organization; therefore good project planning and management
are essential to produce a useful and effective GIS.
1.
Determine the Purpose
of the GIS
Answering some questions
about the project will give a sense of guidance and purpose to the
task about to be undertaken. Clearly defined goals provide a means
of measuring success, so it is important to state these at the start
of the project (Korte, 1997). Questions to consider:
- What are the goals of your GIS?
- What do you want to do with the GIS?
- What are the results you want to
get?
- What are the questions you want our
GIS to answer?
(Carver, Cornelius, Heywood, 1998).
2.
Create a Conceptual
and Physical Data Model
A data model is a term
for the process of identifying all the design elements used in the
construction of a GIS (Carver, Cornelius, Heywood, 1998).
A conceptual model
uses diagrams to show a list of actions that the system must be
able to perform and identifies inputs, such as data sources, and
outputs such as maps. One example of a conceptual model is a flow
diagram. [See Fig 1]
Figure 1: Flow Diagram (SIC, 2002)
A physical model includes additional detail
that describes how to model the spatial entities, their associated
attributes, and the relationships between entities in the computer.
Examples are the raster and vector models discussed further in the
GIS section (Carver, Cornelius, Heywood, 1998).
3.
Needs Assessment/Requirements
Analysis
A needs assessment assembles
information about the potential uses of the proposed system. It
should include the opinions, needs, and requirements of all people
involved in the project. This can be accomplished through interviews
or surveys. The result should include a list of data, processes,
and products that should be accommodated in the final system (Chrisman,
1997).
The requirements analysis
creates a specification for the system. If the organization is
currently using geographic data, the collection, analysis, storage,
presentation, and distribution of it should be reviewed (Chrisman,
1997), (Korte, 1997).
4.
Selecting GIS Software/Hardware
When the search for possible
software and hardware vendors begins, specific specifications should
be incorporated within the organization standard request for bid
(RFB) and request for proposal (RFP) formats. According to The
GIS Book, these should clearly state how the bids and proposals
will be evaluated, the contract terms and conditions, the criteria
for award of the contract, the project schedule, insurance requirements,
penalties, the formats in which bids and proposals must be presented,
and any other instructions to the offers (Korte, 1997).
Once several companies
have matched your minimum requirements, the evaluations of the bids/proposals
must be accomplished. Some questions to consider are:
Software
- What functions does the software
offer?
- Will the functions meet the requirements
of the organization?
- Can it be customized?
-
What training and user support is available?
- Which operating system will be required?
- Will the system meet my needs in
the areas of input, manipulation, analysis, and presentation?
(Carver, Cornelius, Heywood, 1998)
Hardware
- What hardware is necessary for the
application?
- How many computers are needed?
- What storage capacity, memory, speed,
etc, must the computers have?
-
What graphic display devices are needed?
-
What peripherals are needed (printers, plotters, digitizers, scanners,
etc)
- What is the cost?
- How long should it last? (Carver,
Cornelius, Heywood, 1998).
Often companies that meet minimum selection criteria
will perform a capabilities demonstration.
5.
Database Design
The database design
should be a detailed specification defining what objects exist,
their relationships, and how their geometry and attributes interrelate
within the software. The following items should be considered when
developing the specification:
- How the graphic files will be structured
- How non graphic attributes will be
structured
- How file directories will be organized
- How files will be named
- How the project area will be subdivided
geographically
- How the GIS products will be presented
- What security restrictions will be
imposed on file access (Korte, 1997).
If this step is done efficiently, large amounts of
time and money can be saved. It is much less expensive to build
the database correctly the first time than it is to edit the data
later.
6.
Prototype Development
Designing a prototype involves constructing the GIS
database for a small, representative portion of the project area.
The purpose would be to test detailed database design and estimate
costs for data conversion. It is the last opportunity to approve,
cancel, or delay the GIS project before major expenditures are made
(Korte, 1997).
7.
Additional Things to
Consider
- Training – Staff that will use,
operate, and maintain the GIS will need to be trained in its use
and operation
Some of the positions that will be needed
are:
- GIS Manager
- Database Manager
- Cartographer
- Systems Manager
- Programmer
There are several methods to organizing and planning
your GIS project. Some tools that might be useful are:
- GANTT Charts – There are time management
tools that establish when are where, over the life of the project,
a particular task will take place (Carver, Cornelius, Heywood,
1998). [See Fig. 2]
Figure 2: GANTT Chart (SIC,
2002)
- PERT (program evaluation review
techniques) – These are graphical tools for managing a project
by showing how a task depends on the completion of others before
it can be undertaken. One of the main values of a PERT chart
is that it shows which tasks can be undertaken in parallel (Carver,
Cornelius, Heywood, 1998).
Carver, Steve and Sarah Cornelius and Ian
Heywood. An Introduction to
Geographical Information Systems.
Prentice Hall, New Jersey. 1998.
Chrisman, Nicholas. Exploring Geographic
Information Systems. John Wiles and
Sons Inc, New York. 1999.
Korte, George B. The GIS Book. Onward
Press, New Mexico. 1997.
|