Sjöberg, J 2013, 'Numerical analysis, slope design and in situ stress', in PM Dight (ed.), Slope Stability 2013: Proceedings of the 2013 International Symposium on Slope Stability in Open Pit Mining and Civil Engineering, Australian Centre for Geomechanics, Perth, pp. 29-42, https://doi.org/10.36487/ACG_rep/1308_0.2_Sjoberg (https://papers.acg.uwa.edu.au/p/1308_0.2_Sjoberg/) Abstract: This paper presents a brief review of the history of numerical modelling applied to the design of rock slopes, with contributions to previous slope stability symposia used to illustrate the development. The related issue of the effect of in situ (or initial) rock stresses on slope stability is also discussed in the paper. The use of numerical modelling has evolved significantly from the first, precarious, steps in the late 1960s. Currently, modelling is a standard tool in almost all slope design work and highly sophisticated models allow simulation of many (if not all) aspects of slope behaviour. Three-dimensional modelling is common and the use of models that allow explicit crack propagation, is slowly emerging. There is, however, still room for improvements within this field. Many of the existing modelling approaches have not been fully validated, mainly due to the problem of obtaining complete geometrical description and properties for all units within a fractured rock mass. The advance in data collection and generation of input data has not kept pace with the improvements in modelling techniques, and rock mechanics problems remains to a large extent data limited. There is a need for more encompassing data collection and monitoring systems for future modelling improvements. An improved characterisation will require extensive collaborative efforts between engineers, geologists, and geophysicists. In the future, a move towards real-time modelling can also be envisioned. This concept has not yet been fully applied in rock mechanics, but there are examples of applications in other engineering and scientific disciplines, such as weather forecasting and physics modelling, and some ideas are presented.