Complex networks of material flow in manufacturing and logistics: Modeling, analysis, and prediction using stochastic block models

This research addresses the limitations of traditional network models in capturing the complexity and dynamics of real-world social networks. Motivated by the need for a more comprehensive and flexible framework, the study introduces the Hybrid Modern Network Model (HMNM). The HMNM integrates foundational models like the Stochastic Block Model (SBM) and Preferential Attachment with advanced machine learning techniques, including Graph Neural Networks (GNNs), Reinforcement Learning (RL), Hierarchical Random Graphs (HRGs), Generative Adversarial Networks (GANs), and Variational Autoencoders (VAEs). The methods employed involve constructing initial network structures using SBM, simulating network growth through preferential Attachment, learning node embeddings with GNNs, dynamically optimizing network properties using RL, capturing hierarchical community structures with HRGs, controlling degree distributions using GANs, and uncovering latent patterns with VAEs. The empirical illustration of HMNM highlights its effectiveness in providing a more realistic, scalable, and comprehensive analysis of social networks compared to traditional models. Integrating diverse methodologies allows for accurately modeling of network structures, dynamic processes, and latent patterns. In conclusion, the HMNM offers significant advancements in network modeling, providing a robust and flexible framework for analyzing social networks. This model overcomes the limitations of traditional models and delivers deeper insights into the complexities and dynamics of social structures. Future research will optimize the HMNM and explore its applications across various domains. The R programming code used for the network simulations and visualizations is conceptual and demonstrates the HMNM framework. The results and metrics are illustrative placeholders, emphasizing the methodology rather than empirical validation.

Machine Learning Methods for Prediction of Changes in Material Flow Networks

Purpose– The topology of manufacturing systems is specified during the design phase and can afterwards only be adjusted at high expense. The purpose of this paper is to exploit the availability of large-scale data sets in manufacturing by applying measures from complex network theory and from classical performance evaluation to investigate the relation between structure and performance.Design/methodology/approach– The paper develops a manufacturing system network model that is composed of measures from complex network theory. The analysis is based on six company data sets containing up to half a million operation records. The paper uses the network model as a straightforward approach to assess the manufacturing systems and to evaluate the impact of topological measures on fundamental performance figures, e.g., work in process or lateness.Findings– The paper able to show that the manufacturing systems network model is a low-effort approach to quickly assess a manufacturing system. Additionally, the paper demonstrates that manufacturing networks display distinct, non-random network characteristics on a network-wide scale and that the relations between topological and performance key figures are non-linear.Research limitations/implications– The sample consists of six data sets from Germany-based manufacturing companies. As the model is universal, it can easily be applied to further data sets from any industry.Practical implications– The model can be utilized to quickly analyze large data sets without employing classical methods (e.g. simulation studies) which require time-intensive modeling and execution.Originality/value– This paper explores for the first time the application of network figures in manufacturing systems in relation to performance figures by using real data from manufacturing companies.

This paper proposes a reconfiguration optimization method for complex manufacturing systems in terms of reliability analysis. Some misunderstandings of network modelling and reliability analysis for manufacturing systems are clarified. A network model is constructed based on the material flow. Path-dependent and time-dependent characteristics of manufacturing networks are introduced to distinguish from complex network theory. Subsequently, reliability modelling method is proposed, wherein only processing time for individual work system is concluded to compute the reliability. Component importance measure method is given based on former reliability modelling and cost analysis, and it affords the basic input of reconfiguration optimization for manufacturing networks.

Optimization of Artificial Neural Network (ANN) model design for prediction of macroinvertebrates in the Zwalm river basin (Flanders, Belgium)

We explore the relationship between a yield stress fluid flow in a Hele-Shaw cell with irregular walls, and a non-Darcy 2D porous medium flow with limiting pressure gradient. The continuum (Hele-Shaw) flow is a much-studied simplification of a cementing displacement flow model, solved via a variational formulation that leads to a convex streamfunction minimization problem. Our interest is with the analogy between the discretization used for the numerical solution and network flows that model the associated porous media flow, i.e. via the pore-throat approach. A staggered mesh, with pressure nodes at the cell corners and streamfunction nodes at the cell center is used for the continuum problem, which naturally separates into a network representation comprising primal and dual graphs, linking streamfunction and pressure nodes, respectively. We show explicitly how the continuum model defines a network model and vice versa. We develop the variational form of the network flow, including an appropriate (discrete) streamfunction minimization and a discrete version of the principle of virtual work. Two network models are explored, based on different interpretations of the minimization problem. The network flow results are compared with analogous computed continuum flow results in 3 specific geometries. We find that our network model I, which is the most natural interpretation of the continuum model as a network flow using our discretization, generally under-predicts flow rates. This is problematic from the perspective of considering the network flow as an approximation to the porous media or Hele-Shaw flow. Network model II rectifies this situation, via a pressure interpolation method. In our examples we find that the network II flow converges to the continuum flow as the mesh and network are refined. This is not the usual comparison made, as in many pore-throat models the network is fixed according to the underlying pore-space geometry. Despite the differences, both network models have their own advantages and disadvantages. Network model I offers a more natural way of modelling the flow and is easier to apply, for example to complex meshes, e.g. unstructured triangular. Network model II gives the more accurate physical representation of the Hele-Shaw flow. Lastly, we have developed approximate algebraic relationships for the total flow rate as a function of the total applied pressure gradient, both close to the critical onset pressure and for large pressure gradients. These correlations align well with previous findings for Bingham fluid flows in porous media. • Hele-Shaw cell – porous media analogy derived for Herschel-Bulkley fluids. • Discrete pore-throat network models derived. • Discrete principle of virtual work. • Convergence as network is refined. • Improved network model that gives convergence.

Electroencephalogram based communication system for locked in state person using mentally spelled tasks with optimized network model

The paper presents an image-based network model of retinal vasculature taking account of the 3D vascular distribution of the retina. Mouse retinas were prepared using flat-mount technique and vascular images were obtained using confocal microscopy. The vascular morphometric information obtained from confocal images was used for the model development. The network model developed directly represents the vascular geometry of all the large vessels of the arteriolar and venular trees and models the capillaries using uniformly distributed meshes. The vasculatures in different layers of the retina, namely the superficial, intermediate and deep layer, were modelled separately in the network and were linked through connecting vessels. The branching data of the vasculatures was recorded using the method of connectivity matrix of network (the graph theory). Such an approach is able to take into account the detailed vasculature of individual retinas concerned. Using such network model, circulation analyses to predict the spatial distribution of the pressure, flow, hematocrit, apparent viscosity and wall shear stress in the entire retinal vasculature can be carried out.KeywordsMurine retinanetwork topologymorphological datanetwork modelspatial pressureflowratevelocitywall shear stress

In order to solve the problem of excessive parameters and slow computing speed of classical neural networks, this paper uses the powerful parallel computing power of quantum computing to improve the computing speed of classical neural network models, and proposes a hybrid continuous variational quantum neural network (HCVQNN) model that can be used for network intrusion detection. The continuous variable layer of the model is realized through Gaussian gate and non-Gaussian gate, in which Gaussian gate performs operations such as quantum state weight addition and offset term setting, and non-Gaussian gate performs nonlinear operations, thereby improving the overall expression ability of the network model. In addition, aiming at the unbalanced problem of UNSW-NB15 in the network intrusion dataset, this paper proposes to design the algorithm from the feature level and the algorithm level, using the ReliefF algorithm for feature selection at the feature level, and oversampling and undersampling processing at the algorithm level by combining the Borderline-SMOTE algorithm and GMM algorithm. The classification experimental results on the UNSW-NB15 network intrusion dataset show that compared with the other two network models, the HCVQNN network model obtains higher classification accuracy and lower loss function value in both binary classification and multi-classification tasks, and the classification accuracy for minority categories is also improved.

As the metaverse continues to expand, it becomes increasingly critical to have human-centric networks that are both efficient and high-performing to optimize the user experience. Network modeling plays a fundamental role in optimizing and allocating resources efficiently, and configuring networks to satisfy the demands of diverse applications and users. Recently, traditional queuing theory-based approaches to network modeling have given way to machine learning-based methods. These methods rely on vast amounts of data for building precise models. Although high-precision simulators are ubiquitous, data collection is still an expensive and time-consuming process, resulting in a data bottleneck. In this paper, we propose a weakly supervised learning approach to modeling networks for human-centric networking in the metaverse. Specifically, we identify that queuing theory-based labels can be used to design the supervision signal at a very low cost. Therefore, we propose an approach that combines the inaccurate network modeling obtained from queuing theory-based approaches with an efficient and precise network model through only a small amount of simulation data. To make it a reality, we propose a novel neural network model that combines the powerful graph neural network and transformers. Additionally, we propose several additional supervision signals and a training algorithm to build a better network model. Experimental results demonstrate that our approach reduces the burden of data collection while achieving prediction accuracy comparable to results from large amounts of expensive simulation data. Furthermore, our approach exhibits superior generalization ability.

Inhomogeneous random graph models encompass many network models such as stochastic block models and latent position models. We consider the problem of statistical estimation of the matrix of connection probabilities based on the observations of the adjacency matrix of the network. Taking the stochastic block model as an approximation, we construct estimators of network connection probabilities -- the ordinary block constant least squares estimator, and its restricted version. We show that they satisfy oracle inequalities with respect to the block constant oracle. As a consequence, we derive optimal rates of estimation of the probability matrix. Our results cover the important setting of sparse networks. Another consequence consists in establishing upper bounds on the minimax risks for graphon estimation in the $L_2$ norm when the probability matrix is sampled according to a graphon model. These bounds include an additional term accounting for the ``agnostic error induced by the variability of the latent unobserved variables of the graphon model. In this setting, the optimal rates are influenced not only by the bias and variance components as in usual nonparametric problems but also include the third component, which is the agnostic error. The results shed light on the differences between estimation under the empirical loss (the probability matrix estimation) and under the integrated loss (the graphon estimation).

Non-intrusive load disaggregation is designed to estimate the power consumption of each appliance based on the total power of the appliance in the household. Conventional machine learning algorithms cannot accurately extract semantic information from time series data, which motivates us to implement nonintrusive load disaggregation using residual gated recurrent neural networks model (Res-GRU). First, the networks model use multi-scale convolution kernels networks model extract time series data features, and will get multiple map fusions. Secondly, the networks model use residual learning to deepen the network to extract deep load features. Finally, the networks model use the gated recurrent unit to reset and update high level features. In this way, we can get the output power value of the target appliance. The experimental results show that the proposed network model has a good disaggregation effect.

In principle, network models can replicate exactly the microstructure of porous media. In practice, however, network models have been constructed using various assumptions concerning pore structure. This paper presents a network model of a real, disordered porous medium that invokes no assumptions regarding pore structure. The calculated permeability of the model agrees well with measured permeabilities, providing a new and more rigorous confirmation of the validity of the network approach. Several assumptions commonly used in constructing network models are found to be invalid for a random packing of equal spheres. In addition, the model permits quantification of the effect of pore-scale correlation (departure from randomness) upon permeability. The effect is comparable to reported discrepancies between measured permeabilities and predictions of other network models. The implications of this finding are twofold. First, a key assumption of several theories of transport in porous media, namely that pore dimensions are randomly distributed upon a network, may be invalid for real porous systems. Second, efforts both to model and to measure pore-scale correlations could yield more accurate predictions of permeability.

Traditional bridge maintenance approaches often lack the capability to capture interactions among structural elements or assess the vulnerability of freight networks to disruptions such as bridge closures. To overcome these limitations, this study introduces a data-driven bridge maintenance framework that integrates artificial intelligence, geographic information systems, and graph-theory-based network modeling. A comprehensive graph-based network representing Georgia’s National Highway Freight Network was developed using geospatial coordinates and roadway intersection data to evaluate structural and topological criticality at the network level. Structural condition forecasting of individual bridges was conducted using advanced recurrent neural network (RNN) models, including long short-term memory and Gated Recurrent Unit architectures. These models predict future deck condition ratings based on historical element data, average daily truck traffic, age, maximum span length, and the number of main-unit spans. The integrated RNN-driven, graph-theory-based framework uncovers key patterns influencing bridge performance and identifies topological weaknesses that may compromise freight mobility. This analysis enables risk-informed prioritization of maintenance, repair, and replacement strategies, supporting a shift from reactive to predictive decision-making and enhancing the resilience of critical transportation infrastructure. Findings highlight the framework’s utility in evaluating the impacts of disruptive events on bridge closures and network accessibility.

This paper describes the development of an integrated production network model for a major producing field located in the Western Offshore Region of India, the second most prolific oil field in India. The purpose is threefold: a) presentation of a case history describing successful implementation of optimisation methodology in a major gas lifted field, b) outlining the structure and methodology for constructing and calibrating a large integrated production model, and c) illustrating the potential applications in running the completed model in optimisation mode. The Heera Field, currently producing from 119 strings (all on gas lift except one) has been on production since 1984 and with recovery in the region of 20% considerable potential still exists for improving production and recovery. One of the areas and challenges for improvement involves the optimisation of the gas lift performance. The Heera network model described in this paper includes the multi-phase well production, gas lift injection, production processing and gas lift compression facilities that comprise the asset. The development of the network model comprised the following elements: Well models were constructed for all the producing strings using nodal analysis. Outflow performance was calibrated to existing flowing gradient survey (FGS) data and recent production test data. Fluid properties were characterized based on Black Oil correlations using four independent PVT datasets, and tuned to a single correlation across the entire field. Approximately 450 independent pipe/flowline objects were included in the model to represent major pipelines, risers and downcomers. Three major processing platforms, comprising 5 HP separators, MP and LP separators, and 5 gas lift compressors operating in parallel were modelled. Production from neighbouring complexes (Neelam and ICP) was included in the form of "object sources" with overall delivery point at the Uran terminal located onshore. The network model was built, history matched and calibrated to actual field data, within a tolerance of <1% for the measured liquids production. The calibrated model was then used to run a number of optimisation scenarios, in an effort to find ways of improving production and operating efficiency. Optimisation investigations conducted within the framework of this project indicate potential oil uplift of between 6-8% from a combination of well interventions (based on individual well modelling and gas lift diagnostics) and network optimisation (lift gas reallocation), as well as a reduction in total lift gas requirement and compressor horsepower. Besides the scope for production gains and operational improvements, the project also highlights a number of higher level issues relevant to the development and employment of integrated production network models: The successful implementation of the Heera network model is the first of its kind in ONGC's India operations. The technology utilised in this project is capable of modelling large scale complex production networks in a robust and reliable manner. Complexity and size need no longer be barriers to successful implementation of network modelling and optimisation. The technology is easily transferable for implementation by production engineering and operations staff in the field, provided standardised workflows are identified and new work practices implemented to allow improved data management and model maintenance. The maintained integrated production network model provides a platform for analysing and diagnosing a wide variety of production and equipment related issues in the field. Recommendations from the model are currently being implemented in the field. Based on the modelling recommendations, well interventions and lift gas reallocation have so far produced oil production increases in the region of 1700 BOPD.