This paper deals with the application of genetic algorithm (GA) and an adaptive neuro-fuzzy inference scheme (ANFIS), for the prediction of the optimal sizing coefficient of stand-alone photovoltaic supply (SAPVS) systems in remote areas. A database of total solar radiation data for 60 sites in Algeria has been used to determine the iso-reliability curves of a PVS system (C A, C S) for each site. Initially, the GA is used for determining the optimal coefficient (C Aop, C Sop) for each site by minimising the optimal cost (objective function). These coefficients allow the determination of the number of PV modules and the capacity of the battery. Subsequently, an ANFIS is used for the prediction of the optimal coefficient in remote areas based only on geographical coordinates. Therefore, 56 couples of C Aop and C Sop have been used for the training of the network and four couples have been used for testing and validation of the proposed technique. The simulation results have been analysed and compared with the alternative techniques. The technique has been applied and tested for Algeria locations, but it can be generalised for any location in the world.