6. SYSTEM EVALUATION 
Two tests were performed to validate the 
quality of the system. The first test was a 
debugging test that the knowledge engineer was 
performing during the whole stages of 
developing the system. The second test was an 
evaluation behavior of the system and its 
quality. This was accomplished by processing 
many consultation sessions on real problems. 
These sessions were performed by regular 
users. Then the same consultation session on 
the same area is performed manually either by: 
a) a human expert or b) knowledge engineer 
who has ground truth data bout the area of 
testing, without using the system. 
The results of the two consultations (one by the 
system and another by a human expert) were 
compared. The results of four tests are 
presented in Table 3. System interpretation and 
human interpretation are compatible with an 
average confidence of 84%. 
7. CONCLUSIONS 
The developed small system is a powerful one 
that can be used for consultation purposes in 
expansive soils. It can be applied in similar 
region that have the same weather condition as 
that of Tabouk. 
The system was tested and found to be accurate 
as a tool for shale and sandstone identification 
as primary locations for expansive soils. Aerial 
photographs and remote sensing play a great 
role in the identification process. 
The system can be extended to have the 
provision of containing detailed design methods 
of different foundations. The designs should fit 
different cases based on different soil behavior, 
structure sizes, and weather conditions. 
8. REFERENCES 
Al-Garni, A. 1992, Expert System For 
Landform Interpretation Using Terrain Analysis 
57 
(EXLANT), Report No. 423, Ohio State 
University, Graduate School, Columbus. 
Al-garni, A. 1994. The Fourth Dimension in 
Digital Photogrammetry. Photogrammetric 
Engineering and Remote Sensing. Vol. 61, 
No.1, pp. 57-62. 
Al-Jawhra Est. Soil, Foundation, and 
Materials. Tabouk Municipality office, Tabouk 
City, Saudi Arabia. 
Bowerman, R. and D. Glover, 1988. Putting 
Expert System in Practice. Van Nostrand 
Reinhold Company. 
Carbone, G., S. Narumalani, and M. King, 
1996. Photogrammetric Engineering & Remote 
Sensing, Vol. 62, No. 2, pp. 171-179. 
Cracknell, A. and L. Hayes, 1993. Introduction 
to Remote Sensing. Taylor & Francis Ltd, 4 
John St., London WCIN 2ET. 
Curran, P.J., 1992, Principles of Remote 
Sensing, Longman Scientific and Technical. 
Hayes-Roth, F., D., Waterman, and D. Lenat, 
1983. Building Expert System. Addition- 
Wisley Publishing Company Inc. 
Lillesand, T. and R. Keifer, 1987. Remote 
Sensing and Image Interpretation, John Wiley 
and Son. 
Rich, E. and K. Knight, 1991. Artificial 
Intelligence. New York, McGraw-Hill, Inc., 
Second Edition. 
Way, D.S. 1978, Terrain Analysis. Available 
from: 4171 Daventry Road, Upper Arlington, 
OH 43220. U.S.A. 
Mintzer, O., 1989. research in Terrain 
Knowledge Representation for Image 
Interpretation and Terrain Analysis. U.S. Army 
Symposium an A.I. Research for Exploitation 
of Battlefield Environment, EI Paso, Texas, pp. 
271-293. 
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B4. Vienna 1996