Industrial gateways serve as critical data aggregation points within the Industrial Internet of Things (IIoT), enabling seamless data interoperability that empowers enterprises to extract value from equipment data more efficiently. However, their role exposes a fundamental trade-off between computational efficiency and prediction accuracy—a contradiction yet to be fully resolved by existing approaches. The rapid proliferation of IoT devices has led to a corresponding surge in network traffic, posing significant challenges for traffic forecasting methods, while deep learning models like Transformers and GNNs demonstrate high accuracy in traffic prediction, their substantial computational and memory demands hinder effective deployment on resource-constrained industrial gateways, while simple linear models offer relative simplicity, they struggle to effectively capture the complex characteristics of IIoT traffic—which often exhibits high nonlinearity, significant burstiness, and a wide distribution of time scales. The inherent time-varying nature of traffic data further complicates achieving high prediction accuracy. To address these interrelated challenges, we propose the lightweight and theoretically grounded DOA-MSDI-CrossLinear framework, redefining traffic forecasting as a hierarchical decomposition–interaction problem. Unlike existing approaches that simply combine components, we recognize that industrial traffic inherently exhibits scale-dependent temporal correlations requiring explicit decomposition prior to interaction modeling. The Multi-Scale Decomposable Mixing (MDM) module implements this concept through adaptive sequence decomposition, while the Dual Dependency Interaction (DDI) module simultaneously captures dependencies across time and channels. Ultimately, decomposed patterns are fed into an enhanced CrossLinear model to predict flow values for specific future time periods. The Dream Optimization Algorithm (DOA) provides bio-inspired hyperparameter tuning that balances exploration and exploitation—particularly suited for the non-convex optimization scenarios typical in industrial forecasting tasks. Extensive experiments on real industrial IoT datasets thoroughly validate the effectiveness of this approach.

Etching has become a critical step in semiconductor wafer fabrication, and its importance in semiconductor manufacturing highlights the fact that it directly determines the ability of the fab to produce high-process products, as well as the application performance of the chip. While the health of the etcher is a concern, especially for the cooling system, accurately predicting the remaining useful life (RUL) of the etcher cooling system is a critical task. Predictive maintenance (PDM) can be used to monitor the basic condition of the equipment by learning from historical data, and it can help solve the task of RUL prediction. In this paper, we propose the FECAM-WTCN-Informer model, which first obtains a new WTCN structure by inserting wavelet convolution into the TCN, and then combines the discrete cosine transform (DCT) and channel attention mechanism into the temporal neural network (TCN). Multidimensional feature extraction of time series data can be realized, and the processed features are input into the Informer network for prediction. Experimental results show that the method is significantly more accurate in terms of overall prediction performance (MSE, RMSE, and MAE), compared with other state-of-the-art methods, and is suitable for solving the problem of predictive maintenance of etching machine cooling systems.