al 
red 
frame 
k-top 
es) and 
fain 
bytes 
desk- 
peeds 
ral 
ng 
ssing 
ce. By 
'aning 
1e in 
5. 
9? 
mory 
iven 
lated 
that 
format. 
er 
e earth 
ition so 
to 
‚A 
> for 
L0?) or 
er what 
ly 
Itabases 
1 
nmit are 
ISer 
tworks 
stems 
ociated 
T 
out of 
> difficult 
hics and 
nore GIS 
wn in 
ollect 
Reference Ellipsoid 
(Horizontal Datumd 
    
  
Geoid 
(Vertical 
Datum 
VLBI 
Astronomy 
SLR 
LLR 
ISS 
Digital/Analy tical 
Photgrammetry 
Ground Survey 
Spirit Leveling 
GPS 
   
  
Control Segment or 
Geodetic Component 
Projected Surface of 
Reference Ellipsoid 
(Map Pro 
    
Accuracy |Stondards 
  
  
      
  
    
    
  
     
  
  
Data Segment or 
Information Component 
tiond 
  
   
Ground Survey 
      
  
      
    
    
  
Photogrammetry 
LR. Photography 
Digital Photogranmmetr- 
   
Remote Sensi 
(satellite images) 
Radar Images 
Computers or 
ystem Component 
   
   
  
  
     
      
   
      
  
    
  
Storage 
Retrieval 
Manipula tion 
   
Output 
(hard-copy 
devices) 
Information Layers 
Figure 1: Conceptual GIS 
information on part of the surface of the earth, store in a 
computer according to the location with respect to the earth’s 
surface, and restore the information to analyze and 
manipulate for many. applications such as natural resources 
planning, development work, etc. 
Technical problems of GIS such as raster (grid) to vector 
(polygon) conversion and the interchange of digital 
information between different CAMS will influence the 
economic side of the GIS whereas the lack of sufficient 
training will impact GIS implementation and operations. 
Adequate GIS training requires the integration of many 
diverse disciplines taught in a variety of faculties and schools 
such as geodetic sciences (datums, map projections, and 
satellite positioning), management information systems 
(management issues, systems analysis and design, cost 
benefit analyses), survey sciences (conventional surveying, 
legal issues applied to land, photogrammetry, remote 
sensing, and geosciences) and computer science (database 
design, hardware, communications, security, software). 
Figure 1, a scientific and integrated GIS will only be 
possible if due consideration is given to the above mentioned 
components. 
ACCURACY STANDARDS 
A GIS accuracy standard is the degree to which perfection or 
reliability of measurement is attained in a surveying and 
mapping or GIS undertaking. Many organisations, especially 
those in developing countries, do not pay sufficient attention 
to its importance, and many systems presently in use are out 
of date due to rapid development of the technology. One 
must adopt field specifications and methods that will meet 
accuracy requirements and particular mapping standards. 
Accuracy standards are defined as the minimum accuracies 
that are necessary to meet specific objectives [FGCC, 1986, 
1988]. 
The US National Map Accuracy Standard (NMAS) is very 
simple and easy to use because it is expressed in absolute 
terms and without any true scientific statistical concepts. 
NMAS does not provide consistency in the accuracy 
standards in conjunction with mapping from space, and 
computer aided mapping. A new approach for deciding 
statistical map accuracy standards should be designed and 
used for the scientific GIS. For details please refer to 
[Acharya and Bell, 1992]. 
OPTIMIZATION OF GIS 
An optimal GIS has all the constituents of a scientific GIS, 
and is cost effective. The strategy of optimization discussed 
in this paper follows: 
(i) Maximize the accuracy standards: The concept of 
optimization in the past was often to increase the precision, 
accuracy, and reliability, which are interrelated, without 
giving much emphasis on the other factors. In the past, 
given a specific cost, there were not many alternatives, 
because of the limited or expensive technology. 
(ii) Minimize the cost: This should be done by analyzing a 
benefit/cost ratio. The analysis of the benefit/cost ratio is not 
always easy, since many of the economic benefits from the 
GIS output are intangible. 
(iii) Maximize and integrate the product users: The output 
of a GIS is not complete itself but rather is an intermediate 
product. The cost of a GIS can be reduced by increasing 
users, departments and agencies and integrating the systems 
so that duplication is avoided, and costs and data can be 
shared. 
All three strategies should be followed to optimize a GIS 
project. Cost effectiveness analysis is performed by 
analyzing a benefit/cost ratio. The benefit and cost analysis 
are performed separately by assigning benefit and cost 
evaluation factors and then combined to get the ratio. The 
burden of the cost of a GIS project can be substantially 
631