A finite element interpretation is presented of the in situ measurements performed after the occurrence of a large landslide in the vicinity of Ancona, in central Italy. After describing the main geological and geotechnical characteristics of the problem, the results of some finite element elasto-plastic analyses are discussed in which the stability of the slope is investigated under different assumption on the ele- Vation of the water table. It is shown that the analyses based on a standard elastic perfectly plastic material model did not lead to satisfactory results. On the contrary, a simple strain softening model was able to reproduce the main features of the available measurements, including the approximated position of the slip surface.
On presente une interpretation par element finis des mesures ‘in situ’ conduites apres qu'un eboulement etendu eut lieu aux environs de Ancona, dans l'Italie centrale. Apres avoir decrit les principaux caracteristiques geologiques et geotechniques du probleme, on discute les resultats de quelques analyses elasto-plastique par elements finis, ou' la stabilite de la pente est etudiee avec hypotheses differentes sur le niveau de la nappe phreatique. On montre que les analyses que utilisent le modele d'un materiel standard elastique parfaitement plastique ne fournissent pas des resultats satisfaisants. Au contraire, un simple modele de radoucissement dû aux deformations a permis de reproduire le cours principal des mesures a' disposition, y compris la position approximative de la surface de glissement.
Eine Finite Elemente interpretation wird prasentiert von den in situ Messungen, ausgefuhrt nach dem Auftreten einer groben Landverschiebung in der Gegend von Ancona, zentral Italien. Nach einer Beschreibung der geologischen und geotechnischen Haupteigenschafften des Problems, werden die Resultate einiger Finite Elemente elastoplastische Analysen diskutiert in denen die Stabilitat des Hangs untersucht wurde mit Unterschiedlichen Annahmen der Elevation des grundwasserspiegels. Es zeigt sich daß die Analysen welche sich auf ein standard elastisches, perfekt plastisches materialmodell basieren, nicht zu zufriedenstellenderi Resultaten fuhren. 1m Gegenteil, ein einfaches Verzerrungs/Verweichungsmodell war imstande die Haupteigenschafften der verfugbaren Messungen zu reproduzieren, einschließlich der Naherungsposition der verschobenen Oberflache.
The use of numerical techniques for stress analysis can be considered nowadays as a standard practice for the design of complex soil and rock engineering works. In addition to this common use, the numerical techniques may also offer an appreciable help in the interpretation, or back analysis, of the behaviour observed during the construction (according to Terzaghi's observational design method) or during the life of the structure. In soil/rock engineering the term "back analysis" (see e.g. [Cividini & Gioda, 1994]) usually denotes a procedure for finding the values of the parameters of the soil/rock mass that, when introduced in the stress analysis of the problem under examination, lead to results (e.g. displacements, stresses, etc.) as close as possible to the corresponding in situ measurements. In some cases, however, the experimental data may be used not only for "refining" the values of the material constants, but also for "choosing" the most suitable material model, or analysis scheme, among various possible alternatives. In the following, the geological and geotechnical characteristics of the problems are first summarized, considering in particular those aspects which are play a major role on the overall behaviour of the slope. Subsequently, the main features of the adopted elasto-plastic finite element models are illustrated. In particular, some comments are presented on a simple iterative procedure applicable when the shear strength parameters are allowed to decrease with increasing shear deformation.
