Mode: Hybrid. Seats for on-site participants are limited. A confirmation email will be sent to participants who have successfully registered.
“Uncertainties” are ubiquitous in many operation management problems e.g., (multi-stage) inventory control where uncertain demand may lead to extremely high inventory cost. Risk-aware ordering policy thus offers strong reliability guarantee against extremely high cost. We develop a “model-free” method with simple but ingenious structure that could simultaneously estimate decision maker’s risk preference and update the optimal policy. The almost sure convergence and time complexity analysis are derived by revisiting stochastic approximation theory. Numerical studies on inventory control problem valid our theoretical findings. We also introduce several extensions from the basic method and finally point out several future research directions.
Related papers could be found at https://ieeexplore.ieee.org/document/9076320 and https://ojs.aaai.org/index.php/AAAI/article/view/5574
For information, please contact: