Surface defect is a crucial factor affecting the product quality of steel products. Current studies mainly focus on defect recognition and classification using machine vision-based algorithms, which lack the trace of potential causes and the reuse of experiential knowledge. To address this issue, we construct a knowledge graph for steel surface defects by fusing the multi-source and heterogeneous industrial data, including process parameters, chemical compositions, defect images, operation logs and empirical knowledge. A policy-based reinforcement learning approach is developed to solve the path reasoning problem over the industrial knowledge graph in defect detection and diagnosis. The approach employs two agents to explore the path efficiently from opposite directions, utilizes an integrated reward function that comprehensively considers the path direction, path length and entity distance to perform action selection, and adopts the path sharing mechanism and the prior knowledge to update selection policy. Experimental comparisons with the state-of-the-art knowledge reasoning algorithms on two benchmark datasets, NELL-995 and FB15K-237, validate the performance and merits of the proposed method. The effectiveness of the proposed method is also evaluated on a practical steel surface defect dataset, and the results show that our approach performs well in knowledge reasoning on the surface defect graph.Note to Practitioners—The surface quality of products has become a widely concerned focus in manufacturing industries. With the development of industrial IoT and Cyber-physical system technologies, more and more industrial data has been collected, and machine learning-based algorithms have been developed and applied to the recognition of detect defects. However, the algorithms do not take full advantage of the multi-source and heterogeneous defect-related data. On the other hand, it is also difficult to accumulate, inherit and reuse the experts’ knowledge of solving historical cases in the long-term production process. In order to deal with the above obstacles, we apply the knowledge graph for steel surface defect detection. In the proposed approach, a policy-based reinforcement learning algorithm is developed to solve the path reasoning problem over the industrial knowledge graph. To further improve the performance of our algorithm, we employ two agents to explore the path efficiently from opposite directions, utilize an integrated reward function which comprehensively considers the path direction, path length and entity distance to perform action selection, adopt the path sharing mechanism and updated selection policy to reuse the prior knowledge. As a result, our algorithm obtains high precision in knowledge reasoning tasks on two benchmark datasets and a practical steel surface defect dataset compared with some existing algorithms. Hence, it can be readily applied to real surface defect detection problems and facilitates intelligent manufacturing in steel production.

The development of industrial internet and artificial intelligence technologies has accelerated the development of material products. However, decision-making activities in product development are knowledge-intensive and the reusing of knowledge is worthy of studying. In this work, an industrial internet platform is developed to obtain multi-source data from the information systems in steel industry. To mine heterogeneous development knowledge from the tabular data, time series data, pictorial data and textual data, algorithms such as association rule analysis are applied. Metallurgical mechanism knowledge and process knowledge are fused to construct the multimodal knowledge graph through contextual association and ontology modelling. Subgraph matching-based knowledge graph retrieval and reinforcement learning-based knowledge graph reasoning are proposed to efficiently reuse the knowledge in the development of steel products. Practical application validates the effectiveness of the proposed approach in making quick decisions and accelerating the progress of product development.

Camber is a typical asymmetrical defect of slabs in the hot-rolling process. The identification of camber plays a key role in improving the quality of the finished strip. To obtain the shape information of the camber, we select the derivative dynamic time warping (DDTW) distance as the measure of curve similarity. An iterative self-organizing data analysis clustering algorithm integrated with DDTW is proposed to divide the sample into different clusters. A curve template is generated via polynomial fitting in each cluster based on the dynamic mechanism of camber. Subsequently, the dynamic time warping (DTW) and DDTW distances are adopted as the input and a weighted random forest (WRF) algorithm is developed for the camber classification to address the sample imbalance problem. Comparative experiments are conducted on the camber sample collected in the actual factory to test the clustering and classification performances. The experimental results indicate that the error rates (ERRs) of four distance-based methods─Euclidean distance, Pearson distance, DTW, and DDTW are 32.7%, 28.5%, 19.1% and 12.3%, respectively, and ERRs of the four classification algorithms—DDTW-SVM, DDTW-KNN, DDTW-RF, and DDTW-WRF are 9.8%, 5.7%, 4.7%, and 2.1%, respectively, which verifies the superior performance of the proposed method in this article.

Pressures from the market and environment have put forward higher demands to the steel industry, which drive the need to achieve a deeper integration between physical production factory and cyber information system. In this study, a digital twin (DT) platform for hot-rolling production is built based on the whole-process data acquisition, data-driven process modeling, multi-flow coupling modeling and integrated planning and optimization technologies. Surrounding the DT platform, the integrated planning and production scheduling framework for the medium plate plant is introduced. The slab design problem of medium plates is analyzed in detail and solved by Python with an actual production dataset. The developed DT platform has been implemented in a large steel plant, and experimental results show the effectiveness of the proposed model.

In order to respond more quickly to the diverse needs of customers, manufacturing industries have to improve the efficiency of logistics operations in the warehouse. A CPS-based slab yard unmanned crane system is introduced in this paper. The Internet of Things technology realizes the interconnection of logistics equipment in the storage area, and the intelligent scheduling system realizes the collaborative control of the cranes. This paper establishes a mathematical model of the joint storage location and cranes scheduling optimization problem, and integrates the rule-based dynamic scheduling method with a multi-objective optimization algorithm to solve the problem. The actual field application has verified the effectiveness of the system, which reduces the stack transfer rate by 40% and improves the crane utilization by 30%.

Most industrial systems frequently switch their operation modes due to various factors, such as the changing of raw materials, static parameter setpoints, and market demands. To guarantee stable and reliable operation of complex industrial processes under different operation modes, the monitoring strategy has to adapt different operation modes. In addition, different operation modes usually have some common patterns. To address these needs, this article proposes a structure dictionary learning-based method for multimode process monitoring. In order to validate the proposed approach, extensive experiments were conducted on a numerical simulation case, a continuous stirred tank heater (CSTH) process, and an industrial aluminum electrolysis process, in comparison with several stateof-the-art methods. The results show that the proposed method performs better than other conventional methods. Compared with conventional methods, the proposed approach overcomes the assumption that each operation mode of industrial processes should be modeled separately. Therefore, it can effectively detect faulty states. It is worth to mention that the proposed method can not only detect the faulty of the data but also classify the modes of normal data to obtain the operation conditions so as to adopt an appropriate control strategy.

The assembly of customized and complex products entails the collaboration of heterogeneous devices and thus raises the need of effectively planning the assembly process according to the dynamic product and environment information. A cloud CPS-based intelligent assembly system is developed in this paper. Under the framework a product assembly model is introduced to describe the hierarchical relationships and mating features between subassemblies. The integrated assembly knowledge model including product structure, spatial position, mating features, and assembly process is presented. Sensory information collected from the real world and product model together form the assembly context. A two-step assembly knowledge reasoning process is then developed, where similar case matching finds the same or similar product structure from the existing assembly instance library, and priority rules guiding completes the final assembly sequence. A prototype system is developed and the gearbox assembly case validates the effectiveness of the proposed models.

The development of complex products entails the collaborative work of a multidisciplinary team and thus raises the need of effectively supporting knowledge creation and sharing in a collaborative and integrated working environment. A design knowledge model based on hypernetwork is proposed in this paper to facilitate knowledge management and collaborative reasoning in the design and development process. Specifically, the knowledge hypernetwork model is constructed with a designer network, a product network, an issue network and a knowledge unit network. The relationships between various nodes from different networks are identified and defined according to the node properties. On this basis, several topological characteristics of hypernetwork are analyzed and some statistical indicators of the design knowledge hypernetwork are defined. The Bayesian approach is applied to conduct the collaborative reasoning process whereby relevant knowledge units for different design entities are recommended according to the current design tasks and the issues to be resolved. A prototype system is developed and the BIW design case validate the effectiveness of the proposed hypernetwork-based model.

The capability of product family design meets new requirements of customer-driven market nowadays. In order to improve the market competence, enterprises need design a certain product family with better quality, lower cost and less time spending as its best, while these three optimal goals are contradictory in most instances. In this paper, a novel multi-objective collaborative optimization algorithm, Non-dominated Sorting Genetic Algorithm with Hill Climbing algorithm (NSGA-HC), is proposed for product family design, which can trade off local bias of the single-factor algorithm. Secondly, in terms of a better product family designing scheme, a Cycle Subsystem Model with Features is structured, which not only considers product features as an essential factor influencing product performance but also regards each subsystem as individual for product family design. In addition, an industrial field example is demonstrated as a multi-objective optimal problem to validate the proposed model and algorithm. Finally, an experimental comparison shows that this novel algorithm has higher efficiency and avoids local extremum.

Cloud computing technologies have enabled a new paradigm for advanced product development powered by the provision and subscription of computational services in a multi-tenant distributed simulation environment. The description of computational resources and their optimal allocation among tenants with different requirements holds the key to implementing effective software systems for such a paradigm. To address this issue, a systematic framework for monitoring, analyzing and improving system performance is proposed in this research. Specifically, a radial basis function neural network is established to transform simulation tasks with abstract descriptions into specific resource requirements in terms of their quantities and qualities. Additionally, a novel mathematical model is constructed to represent the complex resource allocation process in a multi-tenant computing environment by considering priority-based tenant satisfaction, total computational cost and multi-level load balance. To achieve optimal resource allocation, an improved multi-objective genetic alqorithm is proposed based on the elitist archive and the K -means approaches. As demonstrated in a case study, the proposed framework and methods can effectively support the cloud simulation paradigm and efficiently meet tenants' computational requirements in a distributed environment.

The development of complex products is a complicated multidisciplinary concurrent engineering with complicated task structure and strong technical innovation. The distributed development environment and networked collaborative design model not only bring great challenges to the modeling of complex product development, but also increase the need for risk assessment of the development process. In consideration of the shortcomings that existing risk modeling methods rarely take into account that the multi-layer quality of complex product development structure, the coupling characteristic of concurrent engineering, the simulation involving too many subjective factors and not making joint evaluation for cost risk and schedule risk this paper proposes a multi-layer cost schedule joint risk assessment model based on DSM. Combining with artificial neural networks and the Cholesky factor which can join two independently distributed variables, a DSM-based cost schedule joint simulation method is designed. Finally, this paper gives an introduction to the complex product risk management platform which has been developed and deployed to a relevant enterprise.

3D printing is a kind of important method in the field of rapid prototyping and customized production. In order to achieve the purpose of production integration and product customization of 3D printing, this paper propose a 3D printing cloud integrated platform architecture based on cloud manufacturing. Moreover, how to evaluate and select the services of different distributed 3D printing terminals in cloud platform also requires further in-depth study. To address this issue, a service evaluation model of distributed 3D printing based on cloud manufacturing is designed and proposed in this paper. This model can be used to characterization the evaluation information of each dimension by fuzzy theory, and the evaluation results of the service can be accurately quantified by fuzzy number operation. The illustrative example show that the model and the evaluation method can effectively quantify the service quality, improve the quality and accuracy of services selection.

Modern product development becomes increasingly collaborative and parallel, which raises the need for sharing simulation resources in a multi-tenant simulation environment. However, the service requests and scheduling policies of tenants differ since system resources are shared by tenants. In order to deal with this issue, the framework of monitoring, detecting and scheduling system performance is proposed. As SLA is produced as the contract for negotiations between tenants and the service provider, the WS-Agreement specification is applied and improved to represent the service contents and qualities. Based on the SLA template, a dynamic resource distributing mechanism is realized. The Satisfaction Index is introduced into the scheduling of distributed resources, in which the weighted formula is determined by priorities. A case study has been undertaken and preliminary results show that the proposed solution can effectively support the cloud simulation and efficiently meet tenants' personal requirements.

Schedule and cost factors are largely related to the successful performance of a complex product manufacturing or acquisitions processes. Nonetheless, traditional statistics methods cannot effectively meet the requirement of the complex product management. In this paper, the Weibull distribution with modified least squares estimation (LSE) method is utilized to estimate the integration of schedule and cost. Firstly, the explanation of 3-parameter Weibull probability distribution function for estimation the integrated schedule and cost is demonstrated. In addition, the multi-peaks Weibull distribution is proposed for large complex product modeling consisted of several subsystems. Moreover, a modified LSE method for Weibull parametric analysis is proposed reflecting the latest schedule-cost integration status where the old data will be forgotten with the data accumulation, instead of treating all the data equally. Last but not least, the comparison with other statistics methods demonstrates the effectiveness of method in this research. The illustration using complex system samples also shows that the estimation of schedule and cost integrated based on multi-peaks Weibull distribution is viable and efficient.

The development of product becomes increasingly collaborative and parallel, which makes the need for design reuse more critical. This paper proposes a framework to enable effective development of complex product by supporting the management of design knowledge, which refers to metadata, structural, functional and behavioral information. The multi-level knowledge represented by Base Object Model can be seen as special cases of web services, as BOMs are fundamentally XML documents. Surrounding the effective management of knowledge, the method for the retrieval of multi-level knowledge is developed including the process of retrieving, matching and assembling. Since the knowledge is stored in the form of a tree structure, the depth-first search (DFS) and breadth-first search (BFS) are applied to the retrieval algorithm and similarity is used to calculate the match degree. The framework is implemented in the development of an example missile, which demonstrates that its useful for the re-use of design knowledge.

In order to improve the efficiency of multi-resolution modeling and simulation, Base Object Model (BOM) based multi-resolution model for federated simulation is proposed in this paper. The whole process includes conceptual modeling design, component design, component assembly and federal simulation. BOM-based conceptual models are divided into two categories: basic BOM and the Resolution Related BOM (RR-BOM). Four-Step Modeling method is applied for those two types of BOMs. The completed BOMs are converted to their corresponding component implementations (BCIs). BOM library and BCI library are set up for model reuse. During the components assembly and federated simulation, Dual Engine Simulation (DES) is designed for multi-resolution simulation. Each federate has an Internal Exchange Service Server (IESS) for the internal component management. IESS provide time management, data management, resolution control and other services. The external supporting environment is Run Time Infrastructure (RTI). This method of modeling and simulation has been applied in industrial projects and greatly improved the efficiency for multi-resolution modeling for federated simulation.

Multidisciplinary collaborative simulation is a new paradigm to enhance the analysis of complex engineering systems more accurately in the systematic level, which enabling the interaction of subsystems models constructed by different CAE software tools in a time-synchronized and high fidelity way. This paper presents a modular method to implement the multidisciplinary collaborative simulation for complex engineering systems in a distributed environment. The modular decomposition is based on the configured subsystems in engineering intuition of corresponding disciplines. The global system is described by inter-connections between subsystems with block representation, thus the different simulation modules are interrelated with the corresponding inputs and outputs. The modular simulation approach allows for independent and parallel modeling of each subsystem and has the advantages of a better modularity. A component-based infrastructure is proposed to integrate the distributed, heterogeneous CAE tools. A modular simulation system which can support the integration of distributed simulation models at run-time is developed, involving the implementation technologies of interrelated modules, disciplinary model transformation and simulation time advancement. Modular modeling and simulation of a rolling stock is demonstrated as a typical case study. Numerical results of the modular simulation demonstrate the effectiveness of this approach to multidisciplinary collaborative simulations for complex engineering systems.