Production technology known as additive manufacturing completely deviates from the conventional subtractive method. Due to its unique nature, its application could result in significant Supply Chain (SC) changes and impact the interactions between supply chain participants. This study shows the additive manufacturing applicable in an SC, considers the combination of traditional and additive manufacturing, and redesigns the SC structure. Also, this study aims to reduce operational and traditional costs and provides a new optimization model for changeable multi-level SC. Additive manufacturing is considered both a centralized and decentralized state. Additionally, this paper proposes a new Monte Carlo (MC) method combined with a Machine Learning (MCML) approach to improve the cost uncertainty accuracy compared with simple MC. For validation, the model is tested in a real case and sensitively analyzed regarding changes in the uncertainty and type of manufacturers. The results show that this hybrid model can reduce costs, MCML-based-MPL can increase the uncertainty accuracy in MC, and this model performs considerably better than only one type of traditional or additive manufacturing in SC.

Keywords