Loading...
| Friend's email | |
| Your name | |
| Your email | |
| enter code | |
This page was sent successfuly
- Type of Document: M.Sc. Thesis
- Language: Farsi
- Document No: 54268 (01)
- University: Sharif University of Technology
- Department: Industrial Engineering
- Advisor(s): Modarres, Mohammad
- Abstract:
- One of the lesser-known issues in the modeling literature of assortment optimization issues is the assortment optimization for reusable products, which is an issue in various areas, including offering parking space, renting cloud computing systems resources, and rental of fashion products, etc.In this issue, we are faced with a high diversity product rental store that has a customer in each period. The seller offers a limited set of products depending on the customer demand and the inventory in hand. He can decide to rent the product or leave the system without renting the product. In case of renting the products by the customer, he may return them until the end of this period or keep them for the next periods. Depending on the customer behavior, we have updated the parameters related to the probability of demand. In this research, we use the dynamic programming method to model this problem. To make this model more accurate, we will consider three state parameters that have not been considered simultaneously in the literature. These parameters include the inventory of the seller and the parameters related to the probability of demand, which is updated in each period according to the customer behavior with Beta-Bernoulli learning. This update allows us to learn about demand during the sales period, which makes the model more complex.In this dissertation, it is shown that this model is a model with a degree of exponential complexity. Due to the complexity of this model, we have used a linear approximation method to solve it. In the linear approximation method of this dissertation, starting from a random assortment and recursive solution method, we reach the optimal (or near-optimal) answer for each period. Regarding this linear approximate expressed for a numerical problem, we have also provided a sensitivity analysis of the problem parameters. We also solved a numerical problem to compare the accuracy of this solution algorithm and compared it with the exact solution results. Given the similarity of the proposal presented in both methods, it seems that the developed method is suitable for solving other problems as well
- Keywords:
- Assortment Optimization ; Dynamic Programming ; Demand Learning ; Reusable Products
Digital Object List
-
محتواي کتاب
- view
Bookmark