Search results

The quality of the unstructured mesh has a considerable impact on the stability and accuracy of aerodynamic simulation in computational fluid dynamics (CFD). Typically, engineers spend a significant portion of their time on mesh quality evaluation to ensure a valid, high-quality mesh. The extensive manual interaction and a priori knowledge required to undertake an accurate and timely evaluation process have become a bottleneck in the idealized efficient CFD workflow. This paper aims to introduce a neural network-based quality evaluation approach for unstructured meshes to enable higher efficiency and the level of automation.

The paper investigates the capability of deep neural networks for the quality evaluation of unstructured meshes. For training the network, we build a training dataset for mesh quality learning algorithms. The dataset contains a rich variety of unstructured aircraft meshes with different mesh sizes, densities, cell distribution, growth ratios and cell numbers to ensure its diversity and availability. We also design a neural network, AircraftNet, to learn the effect of mesh quality on the convergent properties of the numerical solutions. The proposed network directly manipulates raw point data in mesh source files rather than passing it to an intermediate data representation. During training, AircraftNet extracts non-linear quality features from high-dimensional data spaces and then automatically predicts the overall quality of the input unstructured mesh.

The paper provides a series of experimental results on GPUs. It shows that AircraftNet is able to effectively analyze the quality-related features like mesh density and distribution from the extracted features and achieve high prediction accuracy on the proposed dataset with even a small number of training runs.

Because of the limited training dataset, the research results may lack generalizability. Therefore, researchers are encouraged to test the proposed propositions further.

The paper publishes a benchmarking dataset for mesh quality learning algorithms and designs a novel neural network approach for unstructured mesh quality evaluation.

The paper aims to propose a generation approach for unstructured surface mesh to speed up mesh generation.

The paper proposes a lightweight interactive generation approach for unstructured surface mesh and presents several key technologies to support this approach.

The experimental results show that the proposed approach is feasible for unstructured meshes and it can accelerate the mesh generation process.

More geometric defects should be covered, and more convenient and efficient interactive means need to be provided.

The proposed approach and key technologies are implemented in NNW-GridStar.UG, which is the unstructured version of the mesh generation software of National Numerical Windtunnel (NNW).

This paper proposes a lightweight interactive approach for unstructured surface mesh generation, which can speed up mesh generation.

Thin cracks on the surface, such as those found in nuclear power plant concrete structures, are difficult to identify because they tend to be thin. This paper aims to design a novel segmentation network, called U-shaped contextual aggregation network (UCAN), for better recognition of weak cracks.

UCAN uses dilated convolutional layers with exponentially changing dilation rates to extract additional contextual features of thin cracks while preserving resolution. Furthermore, this paper has developed a topology-based loss function, called ℓ_cl Dice, which enhances the crack segmentation’s connectivity.

This paper generated five data sets with varying crack widths to evaluate the performance of multiple algorithms. The results show that the UCAN network proposed in this study achieves the highest F1-Score on thinner cracks. Additionally, training the UCAN network with the ℓ_cl Dice improves the F1-Scores compared to using the cross-entropy function alone. These findings demonstrate the effectiveness of the UCAN network and the value of incorporating the ℓ_cl Dice in crack segmentation tasks.

In this paper, an exponentially dilated convolutional layer is constructed to replace the commonly used pooling layer to improve the model receptive field. To address the challenge of preserving fracture connectivity segmentation, this paper introduces ℓ_cl Dice. This design enables UCAN to extract more contextual features while maintaining resolution, thus improving the crack segmentation performance. The proposed method is evaluated using extensive experiments where the results demonstrate the effectiveness of the algorithm.

A digital servitization ecosystem (DSE) is a cooperation model based on the concept of value cocreation. However, capability asymmetry among enterprises can lead to unfair benefit distribution and hinder value cocreation and digital service transformation. This paper aims to investigate the impact of the varying capabilities of enterprises (manufacturers, service providers and digital technology providers) on revenue distribution when these enterprises collaborate on digital servitization transformation. This analysis is performed from an ecosystem perspective to facilitate the stable development of DSEs.

The rise of DSEs has engendered extensive literature, and the distribution of benefits within DSEs is in dire need of new mechanisms to adapt to the new competitive environment. The importance of investment contribution, digital servitization level, digitalization level, risk-taking ability, digital servitization effort level and brand awareness is determined by combining the expert scoring method and the entropy value method to determine different weights for manufacturers, service providers and digital technology providers. The Shapley value is used to design the benefit distribution mechanism for stable cooperation among DSE enterprises, thus providing a more scientific basis for the development of cooperative relationships.

Digital servitization is a collaborative process that involves multienterprise activities, and it is significantly affected by digital servitization level and digitalization level. Moreover, constructing the modified Shapley value benefit distribution mechanism according to the relevant capabilities of digital servitization can promote the stable development of DSEs and value cocreation among members.

The main contributions of this study are as follows: First, it summarizes the stability-influencing factors of DSEs on the basis of empirical and literature research on the demand for enterprise digital servitization capabilities and transformation difficulties, delves deeper into the capability composition and cooperative relationship of DSE members and combines the expert scoring method and the entropy value method to determine the weighting to design the benefit distribution mechanism. Second, it reflects system stability and development by studying the revenue distribution of DSE members, thereby expanding the ecosystem construction and business model transformation of digital servitization in the existing research.

Prefabricated technology is gradually being applied to the construction of subway stations due to its characteristic of mechanization. However, the prefabricated subway station in China is in the initial stage of development, which is prone to construction safety issues. This study aims to evaluate the construction safety risks of prefabricated subway stations in China and formulate corresponding countermeasures to ensure construction safety.

A construction safety risk evaluation index system for the prefabricated subway station was established through literature research and the Delphi method. Furthermore, based on the structure entropy weight method, matter-element theory and evidence theory, a hybrid evaluation model is developed to evaluate the construction safety risks of prefabricated subway stations. The basic probability assignment (BPA) function is obtained using the matter-element theory, the index weight is calculated using the structure entropy weight method to modify the BPA function and the risk evaluation level is determined using the evidence theory. Finally, the reliability and applicability of the evaluation model are verified with a case study of a prefabricated subway station project in China.

The results indicate that the level of construction safety risks in the prefabricated subway station project is relatively low. Man risk, machine risk and method risk are the key factors affecting the overall risk of the project. The evaluation results of the first-level indexes are discussed, and targeted countermeasures are proposed. Therefore, management personnel can deeply understand the construction safety risks of prefabricated subway stations.

This research fills the research gap in the field of construction safety risk assessment of prefabricated subway stations. The methods for construction safety risk assessment are summarized to establish a reliable hybrid evaluation model, laying the foundation for future research. Moreover, the construction safety risk evaluation index system for prefabricated subway stations is proposed, which can be adopted to guide construction safety management.

Megaproject knowledge innovation (MKI) is perceived as a critical strategy for engineering value co-creation and industrial chain upgrading. Ascertaining the impact mechanism of MKI is a crucial initial step towards improving management practices. Within the framework of complex systems in megaprojects, factors exhibit intricate interdependencies. However, the current domain of knowledge has either overlooked or oversimplified this relationship and therefore cannot propose pragmatic and efficacious strategies for enhancing MKI. To close this gap, this study develops a Bayesian network (BN) model aiming to investigate the interdependencies among MKI-related factors and their impact on MKI.

First, this study implements literature review, expert interview and field investigation to identify the influencing factor nodes for the network model development. Second, a Bayesian network was constructed by integrating the expert knowledge with Dempster-Shafer theory. Next, a MKI measurement model was established using 253 training samples. Finally, the factor significance and optimal MKI improvement strategies are identified from the sensitivity analysis and probabilistic reasoning within the BNs.

The results indicate that (1) the BN model exhibits significant reliability and holds promotion and application value in formulating MKI management strategies; (2) knowledge sharing, shared vision and leadership are the key influencing factors of MKI; and (3) simultaneously improving institutional pressure, leadership and knowledge sharing is the most optimal strategy to enhance MKI.

This study innovatively introduced the BN method into the domain of MKI management, providing an appropriate approach for modelling complex relationships among factors and investigate nonlinear influences. The developed model raises megaproject stakeholders’ awareness about factors influencing MKI and presents quantified strategies that increase the likelihood of maximising MKI levels. Its ease of generalisability positions it as a promising decision support tool, facilitating the implementation of sustainable MKI practices.

This study aimed to explore the relationship between supply chain resilience (SCR) and organizational performance (OP), with innovation (INN) serving as a mediator and information sharing (IS) acting as a moderator.

The study comprehensively examined the connections between SCR, OP, INN and IS. An exploratory approach and quantitative methods were employed. The data were collected from small and medium-sized manufacturing enterprises of three cities Xian, Hainan and Guangzhou of China via online questionnaire surveys conducted through Emails and WeChat. SmartPLS-4 was used for data analysis.

The findings indicated that SCR has a positive effect on sustainability efforts. Additionally, INN and effective IS both mediated and moderated this relationship, playing crucial roles in improving sustainability within the supply chain.

The study offered practical insights for businesses to enhance their sustainability efforts. Managers can use these findings to develop strategies that improve SCR, foster INN and encourage effective IS, ultimately resulting in a more sustainable supply chain.

This study enriched the existing knowledge base by investigating the intricate relationships among SCR, OP, INN and IS, all within the context of achieving sustainability. By exploring these elements holistically, the research introduced originality and highlighted effective strategies for sustainable supply chain management.

This study aimed to investigate the relationship between supply chain resilience and organizational performance with innovation as a mediator and information sharing as a moderator.

The study thoroughly explored how supply chain resilience, organizational performance, innovation and information sharing are connected. It used an exploratory approach and quantitative methods. Data were collected from large manufacturing firms through online questionnaire surveys using Google Forms, emails and WhatsApp.

The findings demonstrated that supply chain resilience positively impacts sustainability efforts. Furthermore, leveraging innovation and effective information sharing mediated and moderated the relationship, playing pivotal roles in enhancing sustainability within the supply chain.

The study provided actionable insights for businesses to strengthen their sustainability efforts. Managers could utilize these findings to implement strategies that enhance supply chain resilience, drive innovation and promote effective information sharing, ultimately leading to a more sustainable supply chain.

This study contributed to the existing body of knowledge by examining the complex relationships between supply chain resilience, organizational performance, innovation and information sharing in the context of achieving sustainability. The exploration of these components in a holistic manner added originality to the research and shed light on effective strategies for sustainable supply chain management.

This study aims to examine how degree of servitization and breadth of servitization affect the firm performance, and the moderating roles of corporate social responsibility (CSR) and green innovation.

The hypotheses are tested using panel data regression models from 2010 to 2020 based on 10,539 firm-year observations from 2,362 Chinese listed manufacturers.

Considering the classification of servitization, the results show that both degree and breadth of servitization have positive impacts on firm performance. In addition, CSR strengthens these impacts, while green innovation only strengthens the impact of degree of servitization on firm performance.

This study suggests that manufacturers should obtain greater profits by actively implementing the servitization strategy, and effectively take their social responsibility by achieving value co-creation with stakeholders. Also, they should carry out green innovation, develop green products and provide eco-friendly services.

This study contributes to the servitization literature by deepening the understanding of the value creation effect of services based on the service-oriented logic. Also, this study contributes to the sustainability literature stream by revealing the moderating roles of CSR and green innovation based on the stakeholder theory.