Review of computation methods for stability analysis of slopes

Abstract

In this research a review of various computational methods those are available for slope stability analyses have been performed by considering case studies of three embankment failures and one landslide. These are embankment of Dhaka export processing zone (DEPZ) area in savar, at Gonokbari, Dhaka City Flood Protection Embankment, Dhaka- Narayangonj-Demra (DND) Flood Protection Embankment and Zalesina Landslide, Zagreb, Yugoslavia. Slope stability analysis methods proposed by Fellenius (1936), Bishop (1955), Janbu (1956), Morgenstern and Price (1965) and Spencer (1967) have been reviewed. All these methods are based on limit equilibrium state of the soil at failure. Computer programs have been developed to evaluate the factor of safety. The programs are developed in GWBASIC and termed as modified programs. Both total stress and effective stress types of analyses were performed depending on the available data. A comparison of the results obtained from the various methods of analysis has been carried out. From this comparison, it has been observed that in case of effective stress analysis Fellenius method provides the most conservative value of factor of safety. On the other hand in case of <1>=0 analysis, Bishop's simplified and Fellenius methods, Morgenstern and Price method and Spencer's method yield the same values offactor of safety. The other methods provided the different values of factor of safety. It has been observed that some methods of analyses give the same critical slip surface for a specific condition. An analysis is performed for some selected methods of stability analysis using different procedure for determining the center of critical circle. In this case it is identified whether the same method of analysis give the same critical slip circle or not for use of different procedure for determining the center of critical circle. From the study it has been observed that the same method provides different location of critical circle. An analysis has also been performed by using PCSTABL5M software developed by Purdue University. The slip circle searching technique of PCSTABL5M is different from that of the program developed in this research. In this case Bishop's simplified method and Janbu's simplified methods have been used. The factor of safety obtained from this case has been compared with the factor of safety that obtained from the developed program for the same methods. It has been observed that both the computational techniques provide the same factor of safety for a common slip surface but provide the different location of critical slip surface for the same method of analysis. From the results obtained in this research, it has been observed that in case of Embankment of DEPZ at Savar and Dhaka City flood protection embankment total stress type (<1>=0) of analysis simulated closely the failure condition of the embankments. On the other hand in case of DND embankment the effective stress type of analysis simulated closely the failure condition. In case of Zalesina landslide condition of stability and sliding were checked by determining factor of safety using two different elevations of ground water. It has been observed that FS obtained from Bishop's simplified method and Fellenius method did not satisfy the condition of stability and of sliding for all the selected slip surfaces. Conclusions have been drawn regarding methods of analyses, choice of methods and computational techniques etc. From the study it has been found that Bishop's simplified method is very simple and suitable for stability analysis of circular failure surfaces and Janbu's rigorous method is suitable for noncircular failure surfaces.