The aim of this work is to provide better sales forecasting in the context of a supply chain, by using S&OP model and integrating the time series. We have supposed that difficulties encountered by partners in the forecasts come on one hand from the adaptation of existing statistical models and the other hand from the absence of collaboration and information sharing between partners. Our approach consist in a first step to present the role of forecasts in the relationship between supply chain partner’s and in a second step to recast the process of forecasting relying on a collaborative approach and to develop a scientific model for a monthly sales forecasting.