Stochastic Block Theory is suggested for the observational construction of large underground opening. The probability of forming a removable block by the next excavation can be calculated by taking rock joint distribution into account. The applicability of the theory is examined using the joint trace map.


Theorie Statistique de Roche est suggeree pour la construction experimentale de Ie grand espace souterrain. La probabilite que se former une roche mobile par Ie creusement suivant peut etre calculee en comptant la distribution des fentes. Cette theori est mise en application en utilisant la carte des fentes.


Durch die Observation wird Stochastik Block Theorie fur den Bau von grosser unterirdischer Kaverne geproponiert. Die Moglichkeit, mobilen Block zu formen bei dem nachsten Erdaushub, kann mit dem Denken von der Spalte gerechnt werden. Die Anwendbarkeit von der Theorie wird mit der Spartungmappe exerziert.


For the construction of large underground opening at great depth, the observational construction method is becoming increasingly important to modify the supporting plan of rock masses surrounding the cavern under construction because of the difficulty of estimating the properties of rock masses at great depth before construction. It is well known that the stability of cavern is strongly controlled by joint distribution in the case of hard rock masses. Goodman and Shi proposed "Block Theory" for the estimation of slope stability considering joint distribution. Key block, however, can be found out after all the joint traces of the block appear on the slope surface. In this moment, key block is unstable and can remove out of the slope when occasion demands. Considering above, the authors developed "Stochastic Block Theory" for the effective observational construction of large underground opening. Using Stochastic Block Theory, the stochastic approach for the judgment of key block can be carried out before every joint trace of the block appears on the cavern wall, in other word, in the moment some of the joint traces of the block appear.


It is essentially to predict the existence of key block Which would be formed by the excavation. The geometry of key block on the cavern wall is in the form of finite polygon which is surrounded by some joint traces. Therefore the geometry of the key block which appears partially on the wall, is in the form of polygon which is surrounded by some joint traces and, a "face" or a "bench"(Figure 1). Stochastic Block Theory is applied to such a polygon. The probability of forming a key block by the next excavation can be calculated using rock joint distribution data which are obtained from joint survey on the already-excavated surface.


Figure 2 shows the above mentioned objective polygon on the wall. In general, 2 joint traces(the upper one: joint[A], the lower one: joint[B]) of the objective polygon intersect the face. 2 joint traces are interconnected directly or connected by some joint traces (linking up joints[C]) on the already-excavated wall. The objective polygon is divided into the divergence type and the convergence type on the basis of the directional relationship between joint[A] and joint[B] on the wall[E] which will be exposed by the next excavation as shown in Figure 2. If the objective polygon is a part of the key block, it is necessary that one of the following events.

This content is only available via PDF.
You can access this article if you purchase or spend a download.