deriving parameters or inputs of the model, such as inter-zone distance, node 
table for traffic zones, shortest distance between two nodes, road segment 
length, road linear length, road network length, non linear coefficient, and 
the connectivity matrix between nodes. Some of the intermediate or final 
analysis results are also sent back to the GIS database. The proposed road 
network and its relevent parameters are also evaluated and adjusted. 
2). Urban construction site analysis model 
Another motivation for coupling spatial analytical models with GIS comes from 
GIS-based site selection of urban construction projects. In order to process 
construction permit applications, the planning staff needs the basic 
information (property, planned landuse, building density, infrastructure 
provision capacity, etc.,) which can be done by mapping the desired project 
boundaries and overlaying it with other coverage in GIS. In many cases, a 
suitable construction site should also meet with some other location 
objectives, such as minimizing the maximum distance between facilities and users 
(ninmax model), maxmizing the coverage of the. located facilities (naxcoverage 
model). 
Therefore, the minsum, minmax and maxcoverage models of facility location were 
coupled with GIS to design an urban GIS's construction site analysis model. As 
showed in Fig.2, the spatial entities to be analyzed by the facility location 
models are mapped into spatial objects in GIS, such as the landuse, planned 
landuse, road network, point facilities, etc. Instead of determing the site 
location by pure mathematic calculation, a set of GIS's map-based analysis and 
queries are performed to identify m candidate areas. Then the minmum, minmax 
and maxcoverage models of facility location may be used to selecting thr most 
suitable site location from the m candidate areas via analytic analysis. 
3).Urban terrain landscape analysis model 
While there are many spatial analysis issues which can benefit from the 
coupling of spatial analytical models in some problem domains or some 
application areas. Specific spatial analysis models should be formulated by 
modeling the ‘physically-based spatial process or the human spatial process or 
the human spatial cognitive process. For instance, the terrain landscape. is the 
perceived figure of physical space. During the process of urban or tourist 
landscape analysis is a fundamental but very complicated work. In the request 
of the analysis is a fundamental but very complicated work. In the request of 
the local urban planners, a terrain landscape analysis model was formulated and 
programmed by combining domain specific knowledge of landscape planning process 
with map-based visibility analysis. As showed in Fig.3, the analysis model 
consists of visual field analysis from single and multiple view 
point (s), hierarchical visual field analysis, front-view analysis and visual 
167 
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B4. Vienna 1996