Marl, a transitional soft rock, exhibits complex geotechnical behavior due to its heterogeneous structure, parallel fissures, and variable cementation. The difficulty of obtaining undisturbed samples suitable for laboratory testing makes field-based calibration essential for realistic modeling of excavation performance. This paper presents field observations and numerical back-analysis of a strut-free excavation in marl rock supported by buttress pile walls. The study aims to calibrate the stiffness parameters of the Hardening Soil (HS) model to accurately reproduce the deformation behavior of the marl under actual excavation conditions. Field observations and a monitoring program, including systematic inclinometer measurements, were implemented to provide reliable reference data for model calibration and validation. The excavation was almost entirely in marl, ensuring consistent geotechnical conditions and eliminating the influence of other soil or rock layers that often complicate similar case studies. A three-dimensional finite-element back-analysis was conducted to evaluate the deformation behavior of the self-supporting buttress pile wall during staged excavation and to determine the stiffness parameters that best reproduce the observed field performance. Since both shear strength and stiffness govern wall displacements, shear strength values were fixed based on independent large-scale direct shear tests on intact marl and discontinuity samples from previous studies. Consequently, the analysis focused on identifying stiffness characteristics representative of marl rock under real excavation conditions. The calibrated parameters successfully captured the observed wall and ground deformations, providing a realistic representation of the excavation behavior. The findings offer valuable guidance for similar engineering problems, supporting more reliable prediction of system response in weak and transitional marl formations. Moreover, the results contribute to improved risk management and more informed geotechnical design decisions in future excavation projects under comparable geological and structural conditions.

: This paper presents the influence of complex engineering geological and geotechnical ground conditions on the selection of combined measures for the rehabilitation of landslide on the main road M5. Field and laboratory investigations were conducted to determine the engineering geological and geotechnical conditions of the site. The road is situated on a relatively high embankment within the landslide area, under which a culvert channels a natural stream. The geological substrate is found quite deep, 7 to 8 meters below the surface. Beneath the road, there is a mixture of fill materials of different qualities, indicating that a railway used to run along this route in the past. The slope below the road is relatively steep, approximately 10 meters in height, with a poorly maintained stone wall at its base. The fundamental concept of the proposed solution involves the construction of a support structure comprising piles, a head beam, and a reinforced concrete wall. The designed rehabilitation measures ensure the stability and functionality of the roadway.