1 / 6

Understanding Stochastic Dynamic Programming for Inventory Management Challenges

This handout explores Stochastic Dynamic Programming, focusing on scenarios where external factors, such as unpredictable product demand, impact the decision-making process. Recognizing that the next state is uncertain, it introduces probabilistic models to determine optimal policies at each decision stage. By analyzing an example inventory problem with random demand distributions, we outline the complexities of formulating a production schedule that minimizes total expected costs. This effective modeling technique helps in adapting strategies for real-world applications in inventory management.

terrell
Télécharger la présentation

Understanding Stochastic Dynamic Programming for Inventory Management Challenges

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Dynamic Programming In this handout Stochastic Dynamic Programming

  2. Probabilistic Dynamic Programming So far we have assumed that a specification of the current state and current decision is enough to determine with certainty the state we will end up in the next stage. However, in many practical situations there might be some external factors, like stochastic demand for a product, that might affect the resulting state in the next stage. Thus, it is more realistic to assume that there is a probability distribution for what the next state will be. In this case the model is known as probabilistic dynamic programming.

  3. Probabilistic Dynamic Programming • Suppose we are in state sn of stage n and making a decision xn. • Then the system might go to several different states of stage n+1 based on a probability distribution which is completely determined by the current state and current decision. Namely, the system goes to state i with probability pi (i=1,2,…,S). • We still want to find the optimal policy decision at each stage for each of the possible states. But this time the objective function is an expected value of a random variable. • Because of the probabilistic structure, the recursive relationship in this case is somewhat more complicated.

  4. The basic structure for Probabilistic Dynamic Programming

  5. Inventory problem with random demands • 3 production periods • No inventory at the beginning • Can produce at most 3 units in each period • Can keep at most 2 units in inventory • Set-up cost for each period is 5 The demand is not known in advance. It is given by a probability distribution (see next slide).

  6. Inventory problem with random demands • The demand for each period is given by a probability distribution: • The penalty cost for each unit of unsatisfied demand is 9 • The units exceeding inventory limit are discarded Determine a production schedule to minimize the total expected cost (the DP solution on the board).

More Related