Issue |
E3S Web Conf.
Volume 257, 2021
5th International Workshop on Advances in Energy Science and Environment Engineering (AESEE 2021)
|
|
---|---|---|
Article Number | 03038 | |
Number of page(s) | 6 | |
Section | Environmental Monitoring Repair and Pollution Control | |
DOI | https://doi.org/10.1051/e3sconf/202125703038 | |
Published online | 12 May 2021 |
Supply Chain Scheduling Using Double Deep Time-Series Differential Neural Network
1
School of Electrical Engineering and Intelligentization, Dongguan University of Technology, College of Mechatronics and Control Engineering, Shenzhen University, Shenzhen, China
2
Guangdong Linkcom Technology Co., Ltd, Dongguan, China
3
School of Electrical Engineering and Intelligentization, Dongguan University of Technology, Dongguan, China
4
School of Electrical Engineering and Intelligentization, Dongguan University of Technology, Dongguan, China
b* Corresponding author: 20998353@qq.com
The purpose of supply chain scheduling is to be able to find an optimized plan and strategy so as to optimize the benefits of the entire supply chain. This paper proposes a method for processing tightly coordinated supply chain task scheduling problems based on an improved Double Deep Timing Differential Neural Network (DDTDN) algorithm. The Semi-Markov Decision Process (SMDP) modeling of the state characteristics and action characteristics of the supply chain scheduling problem is realized, so as to transform the task scheduling problem of the tightly coordinated supply chain into a multi-stage decision problem. The deep neural network model can help fit the state value function, and the unique reinforcement learning online evaluation mechanism can realize the selection of the best action strategy combination, and optimize it under the condition of only the stator processing time. Finally, the optimal action strategy group is obtained.
© The Authors, published by EDP Sciences, 2021
This is an Open Access article distributed under the terms of the Creative Commons Attribution License 4.0, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.