Expert system for slope stabilization.
This expert system program is designed to help civil engineers or related users for selection of the possible soil slope stabilizations. The [C.sup.++] program was used as expert shell because of the interfacing capability with the other programs such as stability analysis and GIS programs. The main task of program is consisted of 3 database modules to form knowledge bases and decision making steps as follows.
1. The database of possibly failure modes of soil slopes. In which the data are derived from the technical papers and interviewing the experts in this field.
2. The database for slope stabilization methods. It derived from the expert opinions, the past failure cases and the theoretical considerations.
3. The database of construction cost. The data are from Department of Internal Trades, Ministry of commerce Thailand and other relevant government organizations such as Department of Highways, etc.
The classification of modes of failure is as suggested by VARNES (1978).
The program's flow chart is shown in fig 1. The user can enter the program by the site location of 4 regions of the country according to the Ministry of Interior, as Northern, Southern, North eastern and Central Area regions. Each region is then separated into Provinces, Amphoes and Tambons. The specific locations are entered by the surveying coordinates of the slope such as river bank, embankment, mountainous slope, cut and fill slope. The program proceeds as follows: selection of the region, location, mode of failure, and causes of failure. Then the program will show 3 suitable stabilization methods with safety factor and cost of construction per unit.
[FIGURE 1 OMITTED]
EXPERT SYSTEM RULES AND INFERENCE PROCEDURES
The rules of the expert system were derived by analyzing the knowledgebase of stabilization methods from expert opinions, the past failures cases and theoretical considerations. The 3 suitable stabilization methods were derived from weighted scores by the analytic hierarchy process (AHP). The weighted scores were weighted by sequence of construction, difficulty/facility of construction's maintenance, social's reasons, etc. The schematic diagram for analyzing the suitable inference rule of stabilization method by the failure case is shown in fig 2. The inference procedure of this expert's program is based on forward chaining procedures. The chaining procedures start from a set of conditions and move towards a conclusion.
[FIGURE 2 OMITTED]
RELATIONSHIP OF ENGINEERING SOIL PROPERTIES AND AGRICULTURAL SOIL MAPS
Because the database of engineering soil does not cover all of the country, GIS agricultural soil's map of the Land Development Department (LDD) is used to find the relationship of engineering and agriculture soil. LDD has distinguished soil of Thailand into 62 groups, classified soil about 2-3 meters from ground surface according to soil taxonomy. Each group of agricultural soil consists of established soil series that has specific properties for each location. The soil series is named according to the location. The agricultural GIS map of LDD is separated into tambons, amphoes, and provinces.
The engineering soil's database used in this research are supported by the mobile phone company, DTAC about 95%, soil boring company, STS Engineering about 5%. The soil's bore holes have about 583 bore holes all over Thailand. The Northern area has 135 bore holes. The Southern area has 65 bore holes, The North eastern area has 176 bore holes, and The Central area has 207 bore holes. Almost all soil databases from DTAC have only SPT and depths, no soil classification and other engineering property data. But the database from STS has soil classification and other engineering property data; these data filled engineering soil data completely. Plotting the location of soil bore holes which have the true longitude and latitude position on the agricultural GIS map, the position of bore hole showed the number of soil's group, as shown in fig 3. Then the relationship between engineering soil property and agricultural soil property of that bore hole can be known.
[FIGURE 3 OMITTED]
The essential correlations between the two soils concern soil shear strength and the depth for each soil group. The demonstration of correlations between soil depth and SPT (blows/ft) of soil group No 40 in the North eastern area is shown in fig 4. The strength envelope's equations were fitted by linear equation. By using the lower envelope for calculating the safety factor of stabilization, the stabilized structures were safer than the actual sample. Safety factor of the stabilization method were calculated by the stability program, KUslope. The cost of stabilized structure is derived from estimates the cost of that structure.
[FIGURE 4 OMITTED]
Construction's materials costs were derived from the Department of Internal Trades, Ministry of commerce Thailand and other relevant government organizations such as the Department of Highways or the construction company . The costs of the structures are presented in Baht per unit (meter, square meter, etc).
In Thailand, the growing numbers of landslides caused by human activity and the climatic influence have increased every year. The suitability of stabilization methods depend on the causes of failures. So, the expert system for slope stabilization is a program for providing guidance for selecting a preliminary method for stabilization. Because the soil's shear strength for calculated slope stability of structure uses soil engineering data and soil's agricultural correlations, the user should rechecks soil profiles by boring and recalculated the slope stability of the structure.
The correlations between two soils are rather scattered, the agricultural soil classified by 2-3 meters from ground surface can be used for agricultural purposes only. LDD did not examine the deeper soils and distinguished only 62 groups, so the properties of soils were insufficient to identify any correlations. The correlations between engineering's soil and geological soil may be better than agricultural and engineering's soil because the similarity of processes of soil formation.
The Thailand Research Fund for funding of this research. Assoc Prof Warakorn Mairaing W. for the precious advice. Mr. Dale Kvalheim, English lecturer South East Asia University.
Chearnkiatpradab..B ang Mairaing. W (2002) "Evaluation of soil slope stabilization case studies: Bangpakong river bank and Namkor watered area", Proceeding of the eighth National Convention on civil engineering, Khon Khan, Thailand, vol 2 (In Thai language)
Lee W. Abramson "Slope Stability and Stabilization" John Wiley and Sons Inc
Mairaing. W (2003) "Development of Master Plan for Management of Natural Disasters: Landslides" TRF research, Thailand. (In Thai language)
Varnes, D.J. 1978. "Slope movement types and processes", pp. 11-33. In R.L. Schuster and R.J. Krizek(eds). Landslide Analysis and Control, Transportation Special Report No 176. National Academy faculty of sciences, Washington D.C.
Tansirimongkol W (1999) "AHP the most popular decision method in the word", 1st edition, Graphic and printing, Bangkok (In Thai language).
Kasetsart University, Bangkok, Thailand. Civil Engineering Department, South East Asia University, 19/1 Petchkasem rd. Nongkhem Bangkok 10160 Thailand
|Printer friendly Cite/link Email Feedback|
|Publication:||Geotechnical Engineering for Disaster Mitigation and Rehabilitation|
|Date:||Jan 1, 2005|
|Previous Article:||Formation of the map of shaking of the territory in Almaty region.|
|Next Article:||Triggering and prevention of landslide disasters to cave dwellings in loess region of China.|