PRICE DYNAMICS ON VEGETABLE PRODUCTS BASED ON THE Q-LEARNING ALGORITHM
Vegetables are classified as highly perishable products with a limited shelf life, leading consumers to avoid products approaching expiration. The primary challenge for vegetable sellers is to maximize sales before freshness declines. This study aims to optimize vegetable pricing dynamics to enhance...
Saved in:
| Main Author: | |
|---|---|
| Format: | Final Project |
| Language: | Indonesia |
| Online Access: | https://digilib.itb.ac.id/gdl/view/83271 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Institut Teknologi Bandung |
| Language: | Indonesia |
| Summary: | Vegetables are classified as highly perishable products with a limited shelf life, leading consumers to avoid products approaching expiration. The primary challenge for vegetable sellers is to maximize sales before freshness declines. This study aims to optimize vegetable pricing dynamics to enhance sales and reduce food waste using reinforcement learning with the Q-learning algorithm and its extension, deep Q-learning. Q-learning is an algorithm that discovers the optimal policy, which in this case is the optimal price through experimentation. However, it has limitations due to requiring a large Q-value table. Deep Q-learning overcomes this limitation by using artificial neural networks to estimate Q-values, making it more efficient in handling complex environments. |
|---|