Life cycle assessment (LCA) is a universally recognized methodology for the environmental assessment of products. Its application to the evaluation of complex systems, like agro-systems, for decision making support still faces significant methodological challenges, mainly related to the modelling of indirect consequences of the decisions investigated, e.g. the effects of bioenergy systems on the food chain.For the case of biogas production from maize in Luxembourg, economic modelling approaches, assuming supply-demand equilibrium, are currently investigated in the LUCAS project (CORE2009). However, preliminary results have showed that not always farmers use a fully optimizing approach to farming, as a result of behavioural characteristics like resistance to change and related risks, adversity to innovation, etc. These behavioural criteria are not taken into account in top down economic models, like economic equilibrium models, and they are certainly worthy of further investigation.Multi agent simulation (MAS) is a bottom up technique that can model the heterogeneous and non-optimal nature of decision-making by farmers. It also allows simulating the interactions between different farmers as well as between agents and their environment (institutions, technologies, biophysical context). Several applications of MAS to the management of agro-systems exist in the literature. A couple of pioneering applications of MAS to simple LCAs of energy systems and product design showed the feasibility of the approach and set the basis for further development.MUSA aims at pursuing this line of research by exploring future possible scenarios linked to normative decisions, social models changes, territorial modifications (such as the reduction of agricultural land due to urban sprawl), climatic changes, etc. The approach is thus not seen as an alternative to the economically-based one, but rather as a methodology for analyzing the consequences of future possible evolutions of the agro-system and their consequences on the economic side (even at a longer time horizon, say 2050). The effects of these future evolutions will then be used to estimate the consequences on the market and trigger the consequential modelling in LCA. The expected results are: 1) the development of a practical and very comprehensive methodology of MAS-LCA, and 2) application and validation of the methodology to the case of biogas production from maize in Luxembourg, currently investigated in LUCAS. Dissemination of the results to the scientific community will provide contribution to knowledge development in the field of LCA. Communication to the main players involved in the field of biogas production in Luxembourg (farmers, decision and policy makers, consumers, stakeholders) will prepare the field for further support to policy and decision making in a follow up project.