The sharp, intricate contours of lunar regolith particles hold critical clues to the Moon's geological evolution and inform engineering applications from habitat construction to spacecraft design, making their precise segmentation a task of significant scientific and engineering value. However, this task exposes a weakness in deep learning models known as spectral bias, an inherent tendency to learn smooth, low-frequency functions which causes them to systematically erase the very high-frequency boundary details that are of primary interest. To resolve this conflict, we propose a framework to deeply seek object boundaries. First, we propose High-Frequency Initialized LoRA (HiFi-LoRA) to counteract spectral bias. By initializing the LoRA adaptation matrices as the optimal low-rank approximation of a high-pass filter, it fundamentally enhances the model's high-frequency perception and injects a strong preference for edges. Second, we propose the Wavelet Energy Modulation (WEM) regularizer. It guides the model to learn the intrinsic correlation between contour complexity and mask area, forcing the model to build a geometric understanding of contour morphology upon its high-frequency perception, thereby enabling the generation of boundary details commensurate with the object's scale. Experimentally, we constructed the Lunar Regolith Segmentation Dataset (LRSD), the first large-scale benchmark with expert-annotated contours. Extensive experiments demonstrate that our method sets a new state of the art on this challenging benchmark, not only achieving top performance on regional metrics like mIoU and DSC but, more critically, drastically outperforming existing models on boundary accuracy. This work not only provides a powerful computational tool for lunar science but also offers a robust and synergistic design pattern for other fine-grained segmentation challenges.

The spatial response variation (SRV) is widely employed in frequency-invariant (FI) beamformer design, thanks to the fact that it provides more design degrees of freedom to achieve better FI performance. Recently, the weighted-SRV, a generalized form of SRV, was proposed for the FI beamformer design. It is shown that the weighted-SRV-based design outperforms the SRV-based design with mainlobe ripple and sidelobe level being able to be precisely controlled. However, the approximation error of reference beampattern in the weighted-SRV design may lead to slow convergence or even failure to converge. To address the problem, this paper reformulates the constrained weighted-SRV cost function into an unconstrained form. Under the reformulated cost function, the closed-form solutions of the weighted-SRV's weighting function are theoretically derived, and then an FI-beamformer design approach is proposed. Simulation results demonstrate the superior performance of the proposed approach.

This paper deals with a real-life case study of a Chinese medical device manufacturer. This manufacturer receives orders and delivers products ahead of schedule. A small proportion of orders may not be fulfilled on time due to capacity constraints, changes in demand, etc. These products often have a similar manufacturing process and share the same resources. These products often have a similar manufacturing process and share resources. Items are processed in parallel orders. A multi-stage scheduling decision model is formulated to help production planners make three decisions: (i) whether or not to accept orders; (ii) whether or not to switch demand to produce; (iii) whether or not to adjust production plans. The proposed model takes account of real-time situations in the Production Planning and Control (PPC) system to increase profitability, balance workloads, and avoid delays. Delays in updating demand information and unavailable resources are taken into account. Short- and long-term production plans are synchronized in a multi-level scheduling decision model.

Sustainable scheduling problem has attracted great attention in academia due to the importance of sustain-ability from economical-, environmental- and social-oriented dimensions.). Based on previous studies on the choices of sustainable indicators, system sustainability of production plan-ning and control (PPC) is still perceived to be difficult to be achieved by practitioners. Short-term objectives often conflict with medium- and long-term objectives due to changing de-mands, uncertainties of the shortage of machines and materials and labour's attendance. Previous studies on lot-sizing and scheduling only consider increasing responsiveness to changing or uncertain demands, instead of considering decisions of switching demands. A multi-stage decision-making model for sustainable scheduling in a hybrid flexible flow shop (HFFS) is formulated to take account of whether to accept demands and switch demand from one to the other for achieving system sustainability in this study. This model has been validated by a Chinese medical goods company.