Uncertainty In Rock Mechanics And Rock Engineering - Then And Now

This paper reviews how uncertainty has been dealt with over the past 40 years using the decision making cycle as a frame of reference. The review of the phases of this cycle, as applied to rock mechanics and rock engineering, namely characterization, performance analysis, updating and decision making shows that in each, progress has been made over the past 40 years, although most of the basic concepts existed already in the 1960's. Characterization is the area in which the greatest scientific steps were taken, this phase and performance analysis also gained from the tremendously expanded computational capabilities which allow one to use both more complex and more practically oriented models. The greatest practical advance is without using any doubt in risk based decision making although its acceptance shows interesting differences between different professions.

Dans l'article suivant on utilise le cycle de decision pour illustrer le developpement du traitement des incertitudes en mecanique des roches et en construction en roche dans les quarante annees passees. La revue des phases du cycle de decision montre qu'on a fait du progrès dans chaque phase du cycle de decision, c'est à dire dans la caracterisation, dans l'analyse et la prediction du comportement, dans l'actualisation et dans l'acte de decision, même si les concepts fondamentaux existaient dejà dans les annees 1960. Les plus grands avancements scientifiques se trouvent dans le domaine de la caracterisation tandis que l'analyse predictive profitait du developpement de l'informatique qui a mene à des modèles plus complexes mais aussi plus pratiques. Le plus grand progrès est sans doute dans le domaine de l'acte de decision sur la base des risques. Ce procede est devenu assez commun même s'il y a encore des differences d'acceptation entre differentes professions.

Im vorliegenden Artikel wird der Entscheidungszyklus gebraucht, um darzustellen wie sich der Umgang mit Unsicherheiten im Felsbau wahrend der letzen 40 Jahre entwickelt hat. Der Überblick ueber die Phasen des Entscheidungsprozesses zeigt, dass bei der Anwendung in der Felsmechanik und im Felsbau, in allen Phasen d.h. bei der Charakterisierung, bei der analytischen Vorhersage, beim Aktualisieren und bei der Entscheidungsfindung Fortschritte gemacht wurden, obwohl die grundlegenden Konzepte bereits in den 1960er Jahren existierten. Im Gebiet der Charakterisierung gab es wohl die grössten wissenschaftlichen Entwicklungen, wahrend die analytische Vorhersage vor allem von den Fortschritten der Informatik profitierte, welche komplexere wie auch praktischere Modelle möglich machte. Der grösste Fortschritt ist ganz sicher in der Entscheidungsfindung zu beobachten, wo Risikobetrachtungen gang und gabe sind, obwohl hier noch einige Unterschiede zwischen verschiedenen Berufsgattungen zu finden sind.

Introduction

This paper tries to review how the recognition and treatment of uncertainties in rock mechanics and rock engineering has evolved between the late 1960's and now. It reflects to some extent the work by the author and particularly by colleagues and students. For this reason it is actually not limited to rock mechanics and rock engineering but encompasses geotechnical engineering, in general, as well as engineering geology. Nevertheless, much is “rock oriented” or easily applicable to rocks.

The framework chosen for this review is the well known cycle of “decision making under uncertainty.” (Figure 1). This is quite appropriate for three reasons:

• It was developed in the 1960's


• Although developed for business decisions (Raiffa and Schlaiffer, 1964; Pratt et al., 1965), it is equally applicable to engineering


• It has been the general principle guiding the author's Work

(Figure in full paper)

Exploration and Characterization

Philosophy

Information collection according to Figure 1 takes the form of exploration, and of field and laboratory testing in geotechnical engineering which eventually leads to a characterization of the rock or soil usually in form of qualitative or quantitative descriptions. In essence, information collection is aimed at reducing our uncertainty about the ground characteristics.