Analytical and behavioral aspects of robust data-driven inventory planning

Uncertainty is an essential challenge for any organization. This project focuses on uncertainties in supply chains with highly volatile demand. We formulate a new approach, a powerful, Robust Artificial Intelligence technique, that can deal with the significant data-driven problems in inventory control. Our research consists of two parts. First, we will employ the Robustification of Classification techniques to construct a data-driven uncertainty set. This approach empowers the robust optimization to be entirely data-driven in the absence of predetermined assumptions. Then, we will employ human experiments to examine how humans react to solutions obtained from the data-driven robust optimization approach and what behavioral factors can be incorporated into developing a powerful tool.

Lijia Tan –
Ahmadreza Marandi –
Rob Basten –