Demand forecasting plays a central role in supply chain management and Production Planning and Control (PPC), especially in environments characterized by high uncertainty, volatility, and complexity, such as those associated with Industry 4.0 and PPC 4.0. In this context, traditional statistical models, although widely used, exhibit important limitations when dealing withmultiple seasonal patterns, nonlinear relationships, and large volumes of data and exogenous variables. Grounded in Organizational Information Processing Theory (OIPT), this study assumes that the selection of forecasting models must be aligned with both the level of demand uncertainty and the organization s information processing capacity, seeking a fit between task complexity and analytical sophistication. Using the Design Science Research methodology, a computational artefact is developed in the form of a Decision Support System (DSS) that integrates: (i) a multi-item demand forecasting pipeline combining statistical and Machine Learning models; and (ii) a multicriteria evaluation module that simultaneously accounts for accuracy, bias, stability, and computational performance. The artefact is applied to real data extracted from the ERP system of an industrial company, enabling a model competition across multiple time series at the SKU level. The results show that there is no universallydominant model; instead, different algorithms perform best for specific subsets of items,reinforcing OIPT s contingency logic. The multicriteria assessment makes it possible to adapt model recommendations to distinct organizational priorities (e.g., maximum focus on accuracy, balance between accuracy and computational cost, or emphasis on robustness and stability), providing managers with a practical tool for model selection aligned with the firm s strategy.This study contributes to theory by operationalizing OIPT in the context of demand forecastingand, from a practical standpoint, by delivering a DSS prototype that can be embedded intoSeOP and PPC 4.0 processes, supporting more informed, agile, and strategy consistentdecision-making in complex and digital supply chains.