Artificial intelligence framework for supporting complex decision-making in automotive manufacturing

Globalization has continuously reshaped the structure of consumer industries. In particular, the automotive sector has undergone complex operational changes requiring organizations to rethink logistics and production strategies to maintain competitiveness in a dynamic and interconnected market. Pursuing competitive advantages and added product value depends on more assertive decision-making in increasingly complex production and logistics processes. This study proposes a novel contribution by developing an intelligent framework based on artificial intelligence for demand time series forecasting to support the decision-making process in supply chain operations. The case study focused on the logistics and production operations of a Brazilian automaker operating under the Completely Knocked Down production regime. It aimed to analyze several critical factors, including transit time, parts kit inventory levels, exchange rates, and consumer market behavior. Thus, the purchasing decision model must align with projected future market demand and the most favorable exchange rate. Through this modeling, the framework seeks to optimize decisions regarding what and when to produce or purchase, projecting the best purchasing conditions and facilitating logistical and financial planning. The application of ANNs, therefore, provides critical decision-making support to address the challenges posed by a dynamic and globalized supply chain.