In the quest for sustainable production, manufacturers are increasingly adopting mixed-flow production modes to meet diverse product demands, enabling small-batch production and ensuring swift delivery. A key aspect in this shift is optimizing material distribution scheduling to maintain smooth operations. However, traditional methods frequently encounter challenges due to outdated information tools, irrational task allocation, and suboptimal route planning. Such limitations often result in distribution disarray, unnecessary resource wastage, and general inefficiency, thereby hindering the economic and environmental sustainability of the manufacturing sector. Addressing these challenges, this study introduces a novel dynamic material distribution scheduling optimization model and strategy, leveraging digital twin (DT) technology. This proposed strategy aims to bolster cost-effectiveness while simultaneously supporting environmental sustainability. Our methodology includes developing a route optimization model that minimizes distribution costs, maximizes workstation satisfaction, and reduces carbon emissions. Additionally, we present a cloud–edge computing-based decision framework and explain the DT-based material distribution system’s components and operation. Furthermore, we designed a DT-based dynamic scheduling optimization mechanism, incorporating an improved ant colony optimization algorithm. Numerical experiments based on real data from a partner company revealed that the proposed material distribution scheduling model, strategy, and algorithm can reduce the manufacturer’s distribution operation costs, improve resource utilization, and reduce carbon emissions, thereby enhancing the manufacturer’s economic and environmental sustainability. This research offers innovative insights and perspectives that are crucial for advancing sustainable logistics management and intelligent algorithm design in analogous manufacturing scenarios.

Heterogeneous Information Networks (HINs) contain multiple types of nodes and edges; therefore, they can preserve the semantic information and structure information. Cluster analysis using an HIN has obvious advantages over a transformation into a homogenous information network, which can promote the clustering results of different types of nodes. In our study, we applied a Nonnegative Matrix Tri-Factorization (NMTF) in a cluster analysis of multiple metapaths in HIN. Unlike the parameter estimation method of the probability distribution in previous studies, NMTF can obtain several dependent latent variables simultaneously, and each latent variable in NMTF is associated with the cluster of the corresponding node in the HIN. The method is suited to co-clustering leveraging multiple metapaths in HIN, because NMTF is employed for multiple nonnegative matrix factorizations simultaneously in our study. Experimental results on the real dataset show that the validity and correctness of our method, and the clustering result are better than that of the existing similar clustering algorithm.

As the Internet and Internet of Things (IoT) continue to develop, Heterogeneous Information Networks (HIN) have formed complex interaction relationships among data objects. These relationships are represented by various types of edges (meta-paths) that contain rich semantic information. In the context of IoT data applications, the widespread adoption of Trigger-Action Patterns makes the management and analysis of heterogeneous data particularly important. This study proposes a meta-path-based clustering method for heterogeneous IoT data called I-RankClus, which aims to improve the modeling and analysis efficiency of IoT data. By combining ranking with clustering algorithms, the PageRank algorithm was used to calculate the intraclass influence of objects in the network. The HITS algorithm then transfers the influence to the core objects, thereby optimizing the classification of objects during the clustering process. The I-RankClus algorithm does not process each meta-path individually, but instead integrates multiple meta-paths to enhance the interpretability and clustering performance of the model. The experimental results show that the I-RankClus algorithm can process complex IoT datasets more effectively than traditional clustering methods and provide more accurate clustering outcomes. Furthermore, through a detailed analysis of meta-paths, this study explored the influence and importance of different meta-paths, thereby validating the effectiveness of the algorithm. Overall, the research presented in this paper not only improves the application effects of HINs in IoT data analysis but also provides valuable methods and insights for future network data processing.

The rapid evolution of computer technology and social networks has led to massive data generation through interpersonal communications, necessitating improved methods for information mining and relational analysis in areas such as criminal activity. This paper introduces a Social Network Forensic Analysis model that employs network representation learning to identify and analyze key figures within criminal networks, including leadership structures. The model incorporates traditional web forensics and community algorithms, utilizing concepts such as centrality and similarity measures and integrating the Deepwalk, Line, and Node2vec algorithms to map criminal networks into vector spaces. This maintains node features and structural information that are crucial for the relational analysis. The model refines node relationships through modified random walk sampling, using BFS and DFS, and employs a Continuous Bag-of-Words with Hierarchical Softmax for node vectorization, optimizing the value distribution via the Huffman tree. Hierarchical clustering and distance measures (cosine and Euclidean) were used to identify the key nodes and establish a hierarchy of influence. The findings demonstrate the effectiveness of the model in accurately vectorizing nodes, enhancing inter-node relationship precision, and optimizing clustering, thereby advancing the tools for combating complex criminal networks.

. Manufacturers are increasingly adopting mixed-flow manufacturing models to meet the growing customized, diverse, and dynamic customer demand. In such an environment, material distribution scheduling optimization is vital for smooth operations and is integral to production management. However, manufacturers frequently encounter problems like disordered and delayed material distribution. Traditional scheduling methods suffer from problems like inadequate transparency, delayed decision directives, and suboptimal results, impacting performance. To this end, this study proposes a dynamic material distribution scheduling optimization model and strategy based on digital twin (DT) to address these problems. Firstly, we introduce workstation satisfaction and establish a material distribution path optimization model minimizing total distribution cost while maximizing workstation satisfaction. Subsequently, we present a cloud-edge computing-based decision framework and explain the DT-based material distribution system's components and operation. Furthermore, a dynamic material distribution scheduling optimization mechanism based on DT is designed. By incorporating a savings method and incentive, penalty strategies, improvements are made to the path node selection probabilities and the information pheromone update rules of the traditional ant colony optimization (ACO) algorithm. Finally, a numerical case study, using real data from collaborating enterprises, validates the proposed algorithm and strategy. This research offers valuable insights into logistics management and algorithm design in smart manufacturing environments.

This paper addresses the complex optimization challenge of terminal visit plan formulation faced by Fast-Moving Consumer Goods (FMCG) companies, focusing on reducing operational costs and enhancing employee visit efficiency. Given the large and uneven distribution of terminal stores, relying solely on manual planning or traditional optimization algorithms proves inefficient and costly for both long-term task decomposition and short-term route planning in visit schedules. The study initially tackles the spatial and temporal load balancing issue in the long-term task decomposition phase, where traditional clustering algorithms like k-means fall short. It introduces an improved multi-level KNN-Cluster algorithm, an advanced composite clustering approach that integrates time features with traditional clustering methods, aiming to balance the workload of sales representatives. The algorithm’s effectiveness is verified through experiments against conventional clustering approaches. Addressing the short-term route planning challenge, an NP-hard problem, the study proposes a deep reinforcement learning algorithm based on a policy evaluation network. This algorithm, designed around the Transformer model, models the terminal visit route planning as a sequence-to-sequence problem, analyzing state, action, and reward functions from a reinforcement learning perspective. Experimental results demonstrate the proposed algorithm's capability to effectively solve this problem, outperforming traditional algorithms in terms of solution speed and quality.