Computers in Industry最新文献

The application of blockchain and smart contracts has been widely acknowledged as essential in digitised logistics, offering improved traceability, transparency, and efficiency. However, concerns regarding performance and implementation limitations persist. To demonstrate the challenges regarding the performance and efficiency of blockchain in logistics use cases, this study presents a proof-of-concept model by leveraging the Hyperledger Fabric blockchain network to emulate the shipping logistics process and illustrate the automated and self-executing nature of smart contracts and transactions among various logistics participants by implementing RAFT consensus mechanism. Utilizing Hyperledger Caliper, this study evaluates the performance by systematically adjusting parameters including the number of clients, the number of concurrent transactions, and transaction rates per second. Then nuanced variations in latency, send rate, and throughput are examined. Preliminary findings indicate significant performance impacts related to client numbers and transaction rates per second. When exceeding the processing capacity, the average latency of transactions experiences an exponential increase due to limited resources. Furthermore, different types of operations are compared, with Read operations exhibiting the lowest latency and Update operations displaying the highest latency due to the complex computations and validations involved. Lastly, the latency measures of the LogisticChain network between fixed-rate and linear-rate controllers are compared, highlighting lower latency with fixed-rate controllers. This research contributes to the advancement of knowledge in this field by developing open-source codes specifically tailored for maritime logistics use cases.

Climate change and resource depletion are reshaping economies, compelling governments, society, and businesses to seek solutions that could meet both economic and environmental needs. Due to their relevance to environmental damage, manufacturers are pushed to achieve a sustainable transition in a relatively short time. In this scenario, Industry 4.0 reportedly act as a facilitator of the processes thus leading to the concept of Twin Transition (TT) or digitally-enabled sustainable transition. However, even if literature is aware of the role that I4.0 plays in enhancing sustainability, companies still face a multitude of barriers that hinder the actual implementation of such transition. This paper aims at proposing a new roadmapping methodology to guide manufacturing companies toward TT and link the strategic goals to operations activities. The methodology originates from both an analysis of the barriers faced by manufacturers found in literature and the empirical observations of the authors throughout their research with manufacturing firms. The methodology was implemented in an application case involving 3 independent plants of a multinational company operating in the Food & Beverage sector. The analysis of barriers was performed via a systematic literature review that allowed to identify 39 barriers clustered as Micro (single firm), Meso (network) and Macro (ecosystem). The results show that the methodology applies to single manufacturing plants, and it addresses challenges at micro and meso levels.

This work aims to validate the “minimal AR” authoring approach in a real industrial assembly scenario. It focuses on optimizing visual assets in Augmented Reality (AR) work instructions. The design of AR assembly documentation is influenced by three main variables: work instructions, affordance (dependent on equipment components and operator capabilities), and AR signifiers (combination of visual assets with their properties). In this study, we fixed the instruction complexity while exploring the relationship between affordance and AR signifiers. First, we set up a focus group of 10 experts in AR technical documentation to extract guidelines for the design of minimal AR signifiers for assembly instructions with a variable affordance. Then, we validated these guidelines through an industrial case study involving 34 participants in four assembly tasks. We verified if the candidate minimal AR signifier, obtained using the proposed guidelines, corresponded to the minimal AR signifier established by users. The results showed that in 33% of the cases, users exploited the candidate minimal AR signifier to accomplish the task successfully. Beyond the minimal AR signifier, an additional one conveying the notification about the task success must always be provided to ensure failure by those operators with reduced capabilities. We also found that, in 29% of the cases, users needed less information than the candidate minimal AR signifier due to their higher capabilities. However, as expected, this condition leads users to make more errors than with the candidate minimal AR signifier. Moreover, the study confirms that AR signifiers with redundant information or attractive appearance, such as animated product models, are unnecessary to improve task comprehension. Still, animations could be beneficial in reinforcing understanding when object properties are difficult to detect.

New technologies such as Spatial Augmented Reality (SAR) have created new opportunities for including end users in the early phases of the design process when prototypes are not always available for functional testing. This study investigates the impact of introducing SAR technology on designers and end users working together in co-design sessions. To this end, a multi-modal analysis focusing on cognitive activities and gestural behaviour was conducted. After comparing the traditional setting with the SAR setting (over 44 sessions), the session activity profiles proved to be quite similar. However, the study highlights specific correlations between gestures and cognitive activities, in both environments. Finally, contrary to expectations, free gestures continue to be made in the digitally mediated situation.

This study proposes a methodology for detecting anomalies in the manufacturing industry using a self-supervised representation learning approach based on deep generative models. The challenge arises from the limited availability of data on defective products compared with normal data, leading to degradation in the performance of deep learning models owing to data imbalances. To address this limitation, we propose a process that leverages the Gramian angular field to transform time-series data into images, applies StyleGAN for image augmentation of anomalous data, and utilizes a boosting algorithm for classifier selection in supervised learning. Additionally, we compared the accuracy of the classifier before and after data augmentation. In experimental cases involving CNC milling machine data and wire arc additive manufacturing data, the proposed approach outperformed the approach before augmentation, resulting in improved precision, recall, and F1-score for anomaly detection. Furthermore, Bayesian optimization of the hyperparameters of the boosting algorithm further enhanced the performance metrics. The proposed process effectively addresses the data imbalance problem, and demonstrates its applicability to various manufacturing industries.

Nowadays, manufacturing systems are increasingly embracing the Industry 4.0 paradigm. Therefore, manual and low-standardized manufacturing environments are often digitized through Industrial Internet of Things technologies to quantitatively assess and investigate the role of the human factor from multiple points of view. This approach is commonly known as Operator 4.0. In such a scenario, this manuscript proposes an original digital architecture to monitor the efficiency and the social sustainability of labor-intensive manufacturing job shops. While the anonymous spatio-temporal trajectories of tagged workers are acquired through an ultrawide band radio network, machine learning algorithms autonomously detect the human-process interactions with strategic industrial entities upon developing industrial key performing indicators. The proposed architecture is tested and validated in a real manual manufacturing system. In detail, the performing accuracies of the machine learning-based software provide industrial plant supervisors with several production metrics to identify the hidden weaknesses and bottlenecks of the monitored manufacturing system. Such digital assessment may trigger a re-organization of the considered process to, for instance, enhance the allocation of the material in storage areas while fairly re-balancing the distances traveled by workers for picking activities.

This paper presents a question answering (QA) system that will enable workers from the manufacturing industry to ’hands-free’ request information. This kind of systems, that are broadly used in household context, have started to gain popularity in industrial environments. To develop the system, PDF-based industrial manuals in Spanish have been processed, annotated by experts and used to train the different components of the architecture, i.e. a question classifier and two QA systems. Different metrics have been applied and developed in order to test the performance of the system. The developed architecture obtains high classification and correct response results, demonstrating the viability of these systems in the industry.

Wire and arc additive manufacturing (WAAM) has gradually been applied in industrial applications in recent years due to its low cost, high deposition rate, and high material utilization rate. Anomalies in the WAAM process, such as inclusion, porosity, and lack of fusion, can have unpredictable effects on the quality of the final product. While some studies have investigated anomaly detection methods in the WAAM process, they mainly rely on supervised learning methods that require extensive manual labeling, with less attention paid to unsupervised models. Furthermore, most studies focus on significant anomalies that are rare in actual production, limiting their practical application. This paper proposes a two-stage unsupervised defect detection framework based on online melt pool video data. By considering the motion characteristics of the manufacturing process, a revised threshold method is used to detect anomalies during the WAAM process. Combining machine contextual information, the physical spatial location of defects is further identified and displayed through a human-machine interactive interface. The dataset used in this study is derived from real printing processes of WAAM parts. Compared with baseline methods, the proposed approach significantly improves recall and achieves an F1-score of 86.3% on the test set.

E-commerce customers’ demands for delivery services have become more personalized, diversified, and complex. In this paper, we conduct cluster analysis on the customer demand attributes resulting in a list of attributes including quantitative and qualitative expectations that can be relevant for creating efficient distribution routes taking into account the delivery time and customer satisfaction. A fuzzy clustering optimization method is elaborated for the treatment of above-mentioned customer attributes for distribution management in order to generate efficient delivery strategies. A case study from Shun-Feng (SF) International Express is used to demonstrate the effectiveness and practicability of the proposed method. The obtained results show that both customer satisfaction and the net profit of the enterprise have considerably increased due to an efficient distribution management.

The Digital Twin (DT) constitutes an integration between cyber and physical spaces and has recently become a popular concept in smart manufacturing and Industry 4.0. The related literature provides a DT characterisation and identifies the problem of updating DT models throughout the product life cycle as one of the knowledge gaps. The DT must update its performance by analysing the variable data in real time of the physical asset, whose behaviour is constantly changing over time. The automatic update process involves a data quality problem, i.e., ensuring that the captured values do not come from measurement or provoked errors. In this work, a novel methodology has been proposed to achieve data quality in the interconnection between digital and physical spaces. The methodology is applied to a real case study using the DT of a real solar cooling plant, acting as a learning decision support system that ensures the quality of the data during the update of the DT. The implementation of the methodology integrates a neurofuzzy system to detect failures and a recurrent neural network to predict the size of the errors. Experiments were carried out using historical plant data that showed great results in terms of detection and prediction accuracy, demonstrating the feasibility of applying the methodology in terms of computation time.