ORIGINAL PAPER Design of large cuttings in jointed rock M. Gasc-Barbier Æ A. Ballion Æ D. Virely Received: 17 February 2007 / Accepted: 17 November 2007 / Published online: 19 February 2008 Ó Springer-Verlag 2008 Abstract The article presents a stability study on a very high cutting in a granitic gneiss, as part of the construction of a bypass in the French Pyrenees. The preliminary study highlighted questions as to the most appropriate method of designing the rock cuts, as such methods are not stan- dardized in France. A temporary stop in the roadworks allowed a second study to be carried out based on a statistical analysis of the discontinuities of the site and better-developed modeling tools, including Resoblok. A comparative analysis of these methods and their results in terms of bolt sizing is presented, taking into account the difficulty of use of each method. Keywords Discontinuity  Rock mass stability  Bolting  Design  Structural model Re ´sume ´ l’article pre ´sente une e ´tude de stabilite ´ re ´alise ´e sur des de ´blais de grande hauteur en cours de re ´alisation dans un massif gneissique, pour la cre ´ation d’une de ´viation dans les Pyre ´ne ´es franc ¸aises. Dans un premier temps, l’e ´tude pre ´alable au chantier a ` conduit les ge ´otechniciens a ` s’interroger sur les me ´thodes de dimensionnement de tels talus rocheux, me ´thodes non norme ´es a ` l’heure actuelle. Dans un second temps, et gra ˆce a ` l’arre ˆt temporaire du dit chantier, une seconde e ´tude a e ´te ´ re ´alise ´e en se basant sur une analyse statistique plus fine des discontinuite ´s du site et sur des moyens de mode ´lisation plus de ´veloppe ´s. Une analyse compare ´e de ces me ´thodes et de leurs re ´sultats en terme de dimensionnement du clouage est propose ´e, tout en tenant compte de leurs lourdeurs respectives. Mots cle ´s Discontinuite ´s  Stabilite ´ de parois rocheuses  Boulonnage  Dimensionnement  Mode ´lisation structurale Introduction A variety of tools are available for studying rock mass stability depending on the standpoint of the user. In an engineering context, they should be efficient and give results that are not open to discussion from a technical or legal point of view should problems arise during the road works. Furthermore, the tools should be easy to use and not require too much time to provide good results. The results should be as general as possible so that they can be applied to large parts of the site. Researchers can use in situ measurements, laboratory tests or software that is still in development to help them understand the physical or mechanical phenomena. Results are not considered as absolute but more as indications to help understand the most important phenomena. Software packages are often rather complicated and require time and precise site information to give useful results. As a con- sequence, the results are often very local and it can be hard to extrapolate to a larger model (Alfonsi and Fouche ´ 2005). For studies on slope instability, software can be based on continuous or discontinuous media mechanics depending on the purpose of the model and the site under consider- ation (soil or rock, jointed or not, etc.). The importance of predicting the bulk of rock that may fall after cutting can easily be understood, as it will indicate the type of protection that has to be used. If the unstable blocks are small, wire mesh is a relatively easy and cheap solution. If the volume of the unstable blocks is greater, bolts are necessary. Passive bolts allow a composite mass to be created while active bolts increase the shear M. Gasc-Barbier (&)  A. Ballion  D. Virely Laboratoire Re ´gional des Ponts et Chausse ´es de Toulouse, 1 avenue du colonel Roche, 31400 Toulouse, France e-mail: Muriel.gasc@equipement.gouv.fr 123 Bull Eng Geol Environ (2008) 67:227–235 DOI 10.1007/s10064-008-0127-4