.png){kind=icon}
.png){kind=icon}
Return
Fuzzy-neural cloud-edge computing application for internal due date assignment: wafer fabrication as an example
.png){kind=icon}
PRE
AI
C
W
DOI：10.1007/s13748-026-00440-0.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Internal due date assignment is crucial for customer relationship management and helps build company competitiveness and sustainability. This study proposes a fuzzy-neural cloud-edge computing application for internal due date assignment, which is the first attempt at hybridizing cloud and edge computing for supporting such managerial activities. The fuzzy-neural cloud-edge computing application first breaks down a possible order into lots and virtually release them based on daily quotas. Subsequently, a fuzzy deep learning (FDL) model is created on the cloud server to estimate the output time of these lots to obtain the completion time of the possible order, so as to assign the internal due date. The FDL consists of a sparse autoencoder for handling big data and protecting data security, and a fuzzy feedforward neural network (FFNN) for estimating lot output time and establishing the upper bound. To facilitate the communication and negotiation of the internal due date assignment mechanism and result, a selective random forest is constructed to approximate the FFNN. The fuzzy-neural cloud-edge computing approach for internal due date assignment has been applied to a wafer fabrication case. According to the experimental results, the solution provided by the fuzzy-neural cloud-edge computing application was more secure, efficient, and attractive than those of common practice and cloud computing counterparts. In addition, the fuzzy-neural cloud-edge computing application also explained and communicated the internal due date assignment mechanism and result more precisely and intuitively.
Keywords:
Cloud edge computing
Manufacturing
Internal due date assignment
Deep learning
Fuzzy logic
Journal
P
IF:
2.4
Papers:
44
Citations:
0
Organization
Cited Papers