Uncertainty about the future and orebody should significantly impact evaluation of mining projects. Scenario planning, widely used in other industries, for example in the petroleum industry, can provide organisations with an opportunity to explore the consequences of uncertainty on their future plans. By developing plausible scenarios, scenario planning assists decision makers by allowing them to make systematic and possibly effective decisions for the future. This paper reviews existing mine project evaluation in the context of a new framework called scenario-based project evaluation (SBPE) (Vann et al , 2012) that addresses many shortcomings of existing approaches. SBPE uses conditional simulations (CS) models as inputs which are evaluated against future alternative project pathways (alternatives'). These alternatives are evaluated using multiple plausible images of the deposit (CS realisations) which capture realistic variability and as an ensemble, map uncertainty of various orebody attributes at fine scale. The alternatives can involve different mining, stockpiling, blending and processing strategies. The ultimate output is a (stochastic) set of cash flows for each alternative from which financial measures, such as NPV, can be derived. Alternatives can then be explored in the presence of against external scenarios such as changes in prices, costs, and taxation. The obvious problem with such data rich results is knowing which scenarios or group of scenarios are optimal.An innovative approach to robust stochastic optimisation (RSO) suited to the SBPE framework is presented here for the first time. This approach to RSO aims to deal with the large number of alternatives generated from SPBE. The goal of applying the RSO approach to SBPE is to identify robust strategic alternatives (project configurations) which have the flexibility to deal with variability and uncertainty of geological and processing inputs such as density, grade and process performance. By applying the proposed RSO method, it is possible to deal with uncertainty and identify the best project configuration in the face of a large number of envisaged scenarios. A case study is provided, in which we have tested the application of the proposed method and compared the result with conventional methods. An important conclusion from this research is that selecting the value of the penalty coefficients in the RSO approach proposed is (as is all optimisation) an exercise of trade-offs; a higher net present value (NPV) is generally associated with a higher risk and a lower risk is associated with a lower NPV. Indeed, this is a strong argument for considering the RSO approach outlined here over more conventional approaches.CITATION:Moayer, S, Vann, J, Coward, S, Jackson, S, Bye, A and Wolff, R, 2012. An innovative approach to robust optimisation for scenario-based mineral project evaluation, in Proceedings Project Evaluation 2012 , pp 141-148 (The Australasian Institute of Mining and Metallurgy: Melbourne).