Abstracts:The industrial scenario is undergoing exponential changes, mainly due to the different technologies that emerge quickly and the ever increasing demand. As a consequence, the number of processing devices and systems in the industries’ architectures is also increasing. Entities connectivity, physical/virtual joint functioning, interactivity, interoperability, self-organization, smart decision making, among other factors are fundamental to foster Industry 4.0 (I4.0) potential. We believe that Cyber Physical System (CPS) and Industrial Internet of Things (IIoT) will have a major role in the emerging I4.0. In this context, researchers and experts from major factories are exploring these technologies in order to keep up with this digital transformation, developing IIoT systems and CPS architectures capable of connecting network devices from different information and communications technologies (ICT) systems, virtualizing the companies’ assets and integrating them with other manufacturing sectors and companies. This article performs a survey covering the main CPS architecture models available in the industrial environment, emphasizing their key characteristics and technologies, as well as the correlations among them, pointing objectives, advantages and contribution for the IIoT introduction in I4.0. It also provides a literature review covering projects from CPSs and IIoT point-of-view, identifying main technologies employed in current state-of-the-art and how they can meet the I4.0 key features of vertical and horizontal industrial integration. Finally, the article points requirements for current and future challenges, limitations, gaps and necessary changes in the CPS architectures in order to improve and introduce them in the I4.0 scenario.

Abstracts:With the popularity of service-oriented manufacturing mode, the customer quantities of the online manufacturing service platforms are growing exponentially. To improve the user-friendliness and convenience of online platforms, the personalized service recommendation for different customer requirement is an effective means. However, since manufacturing services usually appear in the form of composite services, existing Web service-based personalized recommendation technologies are difficult to be applied effectively. Therefore, this paper proposes a novel hybrid algorithm to address the personalized recommendation for manufacturing service composition (MSC). The algorithm solves the insufficient individualization defect of MSC optimization by comprehensively considering the QoS objective attributes and customer preference attributes. First, a Clustering-based Collaborative Filtering (CCF) algorithm is proposed to quantify the customer preference attributes. Second, an improved Personalization-oriented third generation Non-dominated Sorting Genetic Algorithm (PoNSGA-III) is presented for the multi-attribute MSC optimization. Finally, the hybrid algorithm recommends the most suitable solutions for the target customer through the ranking of customer preference attributes. A detailed case study is designed to demonstrate the performance and practicability of the proposed recommendation algorithm.

Abstracts:Product design information represents not only the carrier of design but also the significant digital assets of businesses. At present, manufacturing is facing an environment with mass, fragmented, real-time, and multi-scene digital information in the process of product design. To improve the availability of information resources and the efficiency of information reuse as well as to achieve the sharing of production means, the expression of product design information should be optimized in information storage. In addition, a new ecosystem should be built for information increments, making the participants of every link in the process of product design become the contributors of information. Considering the unique aspects of design group individuals, this paper builds a knowledge transfer model that capitalizes on hypercycle theory and proposes the concept of modularizing information carriers and information processing methods. It comprehensively analyses the expression methodology and organizational attributes of product design information and proposes a knowledge transfer carrier model constructed via discretized fragmented semantic information, which is concretely implemented as informative product file labelling. By combining personnel, information carriers and information dissemination networks, this paper provides the functionality and architecture needed to build an information processing platform using social networking software (SNS). Some application scenarios are described by using this processing method for production information; the development process for an automotive air filter is shown as an example. The results of this study suggest that the proposed method is conducive to improving the expressions of product design information and the interactions among participants. The simplicity of the labelling process and the intuitive label content greatly reduce the usability threshold and the losses caused by information gaps, creating a precondition for many types of people to fully participate in the product design information knowledge transfer cycle.

Abstracts:Today, due to increased market competition, the integration of production and distribution decisions into the supply chain leads to efficiency improvements. Therefore, production routing models have been developed to optimize production and distribution, simultaneously. In recent years, since, the product life cycle has become shorter than in the past, product return policies with fast response times, emphasis on return management, deformation and restorage of finished goods have become very significant. In this paper, the multi-product production routing problem with startup costs and environmental considerations has been studied. Furthermore, reverse logistics and remanufacturing decisions have been integrated. After modeling the problem as mixed-integer linear programming, due to its NP-hardness and the successful application of the Lagrangian Relaxation algorithm (LR) in solving complex supply chain problems, this algorithm has been chosen as the solution method. After applying the standard LR algorithm and the improved LR algorithm that its subgradient optimization method was modified, as a heuristic algorithm, the feasiblizer algorithm is also proposed for feasibilization of the solution obtained from the LR algorithms. To validate the model and solution method, firstly, test problems are solved by GAMS, and then the proposed algorithm is applied to test problems. The numerical results show the good performance of the LR algorithm in medium-size test problems. Finally, based on the computational experiments, managerial insights on the problem have been provided.

Abstracts:Ultrasonic welding is a well-known process for joining thermoplastics and has recently been introduced for joining carbon fiber reinforced polymer (CFRP) composite materials in the automotive industry. As a new joining method for CFRP materials, an understanding of the impact of the welding process on weld attributes and joint performance such as lap-shear strength is needed, as are methods to effectively classify weld quality. This paper investigates the relationship between joint performance and weld energy in ultrasonic welding of injection molded thin short-fiber CFRP sheets. Weld quality classes for training a generalized algorithm are determined from welded joint lap-shear strength and the microstructure of the weld zone. A simple and efficient method for feature selection is proposed to screen the most significant features for predicting from multiple weld quality classes. Several feature selection and weld quality classification methods were compared. A Bayesian Regularized Neural Network (BRNN) was found to be more accurate and robust when classifying weld quality in ultrasonic composite welding than the previously proposed methods of support vector machine (SVM), k-nearest neighbors (kNN), and linear discriminant analysis (LDA).

Abstracts:Facility layout problems focus on the assignment of departments into the facility layout by considering the minimization of total costs. A well-designed facility result in efficient material flows in transportation. For this reason, the problem should be considered with the changes in demands that cause different material flows between departments. While the aim of the static facility layout problem is to obtain the optimal facility layout, the dynamic facility layout problem includes the multi-periods dealing with the changes in material flows between departments. Furthermore, the material handling devices may affect the facility layout decisions although it is omitted in general. In this study, three different facility layout problems concern static, dynamic and flexible bay structure and material handling device assignment problems are considered simultaneously. Three mixed integer programming models are proposed for the addressed problems with integrated decisions. The proposed models are compared with a two-stage approach composed of solving the classical facility layout problems first, then solving material handling devices assignment problems. The comparative results show that the integrated decision affects the facility layout and decreases the total costs. Eleven test instances are considered and the integrated model provides cost efficiency for these test instances.

Abstracts:An inventory control problem of a hybrid manufacturing/remanufacturing system under stochastic demand is studied. The paper proposes a successive substitution strategy between prime, budget and low budget segments of a market. Meaning a manufactured product may substitute for a remanufactured product at a discounted rate. The success of this proposal proportionally is dependent on the discount rate. Similarly, a once-remanufactured product may substitute for a twice-remanufactured product. In this proposal, a sales-dependent return is adopted to enhance returns prediction. This tactic utilizes a virtual stock of previous period sales instead of a used product stock. In such arrangements, only a fraction of customers accept keeping and returning their used product at the end of the product's lifetime. This is a push system because as soon as the company makes an order for remanufacturing, a core will be returned by a ready/potential customer and instantly will be pushed into the remanufacturing process. The system is formulated by a bi-objective Markov decision process considering profit and customer satisfaction criteria to provide optimal policy in the form of a lookup table. The heuristic policy is then characterized by utilizing five control parameters. This is a state-dependent threshold policy with a small deviation from the optimality. Then the effects of thresholds are investigated applying DOE and ANOVA. Finally, the performance of the successive substitution strategy using real data from the tire industry is examined. The results show that profitability, serviceability and core acquisition can be improved by using the proposed strategy.

Abstracts:In the last years, the increase of data availability together with enhanced computation capabilities empowered researchers to conceive production planning and control methods with real-time inputs. Literature is rich with techniques for using simulation to take production planning and control decisions online. However, it is generally impractical to test these approaches on real systems, and experiments on digital instances are limited because they do not capture the physical aspects. This work proposes to test Real-time Simulation approaches using lab-scale models of manufacturing systems and a software architecture aligned with industrial standards. Such models allow to reproduce material flows and the production control logic of real factory environments. By exploiting this setting to test new approaches and tools, it is possible to increase their own achievable Technology Readiness Level (TRL). The laboratory has been used to set a real-time rescheduling problem on a Flexible Manufacturing System (FMS) model. The test involves simulation models aligned with the current system state for the online identification and implementation of a production scheduling rule that decreases the expected makespan. The results testify that the proposed lab-scale models can be used successfully to test production planning and control approaches.

Abstracts:Remaining useful life (RUL) prediction plays an important role in increasing the availability and productivity of industrial manufacturing systems. This paper proposes a joint classification-regression scheme for multi-stage RUL prediction. First, the time domain and frequency domain features are extracted from various types of raw sensory data (e.g., acoustic, current, vibration and temperature) to constitute the training data set. Second, the system health stage is classified based on the trained model and real-time sensory data. Third, we perform stage-level RUL prediction with regression algorithm to estimate overall useful life. Distinct from the existing RUL estimation algorithms, the proposed multi-stage remaining useful life (MS-RUL) prediction effectively integrates the machine/deep learning based classification and regression to improve overall estimation accuracy. We conduct the performance evaluation with sensory data from real manufacturing systems. Experimental results demonstrate that the proposed MS-RUL achieves approximately 6.5% accuracy improvements over the state-of-the-art algorithms in the RUL prediction.

Abstracts:Key performance indicators (KPIs) modeling and control is important for efficient design and operation of complex manufacturing production systems. This paper proposes to implement the KPI control based on KPI modeling and stochastic optimization. The KPI relationship is first approximated using ordered block model and pair-copula construction (OBM-PCC) model, which is a non-parametric model that facilitates a flexible surrogate of the KPI relationship. Then, the KPI control is framed into a stochastic optimization problem, where the randomness in the cost function depends on the decision variables. To solve this stochastic optimization problem, the standard uniform distribution is employed to link the OBM-PCC model and the cost function to transform the problem into an ordinary stochastic optimization problem. The proposed method is efficient in KPI control and the performance is robust to the cost function. Extensive numerical studies and comparisons, together with a case study, are presented to demonstrate the effectiveness of the proposed KPI control framework.