Search results

The heterogeneous character of Industry 4.0 opens opportunities for studies to understand the difficulties and challenges found in the transformation process of manufacturers. This article aims to present a critical analysis of the modernization process of an Industry 3.0 automated cell into a fully autonomous cell of Industry 4.0. The objective is to elucidate the difficulties found in this transition process and the possible ways to overcome the challenges, focusing on the management perspective.

For this, the needed steps for the technology transition were defined and the main I4.0 enabling technologies were applied, such as the application of machine learning algorithms to control quality parameters in milling.

The main challenges found were related to the obsolescence of the equipment present in the cell, challenges in data integration and communication protocols, in addition to the training of people who work actively in the project team. The difficulties faced were discussed based on similar studies in the literature and possible solutions for each challenge.

This understanding of possible barriers in the modernization process, and the step-by-step defined for this transition, can be important references for professionals working in manufacturing industries and researchers who aim to deepen their studies in this important and disruptive stage of world industrialization.

This paper aims to propose a lightweight, high-accuracy object detection model designed to enhance seam tracking quality under strong arcs and splashes condition. Simultaneously, the model aims to reduce computational costs.

The lightweight model is constructed based on Single Shot Multibox Detector (SSD). First, a neural architecture search method based on meta-learning and genetic algorithm is introduced to optimize pruning strategy, reducing human intervention and improving efficiency. Additionally, the Alternating Direction Method of Multipliers (ADMM) is used to perform structural pruning on SSD, effectively compressing the model with minimal loss of accuracy.

Compared to state-of-the-art models, this method better balances feature extraction accuracy and inference speed. Furthermore, seam tracking experiments on this welding robot experimental platform demonstrate that the proposed method exhibits excellent accuracy and robustness in practical applications.

This paper presents an innovative approach that combines ADMM structural pruning and meta-learning-based neural architecture search to significantly enhance the efficiency and performance of the SSD network. This method reduces computational cost while ensuring high detection accuracy, providing a reliable solution for welding robot laser vision systems in practical applications.

With the aid of bibliometric analysis, this study aims to show the state-of-the-art of research on the digital divide and identifies new areas for further investigation.

Performance analysis and science mapping were used in the study to analyse a sample of 3,571 studies that were published between 2018 and 2022. The “Title-Keyword-Abstract” search option was used to collect the anticipated publications data from the Scopus database. The gathered data were analysed using the common bibliometric indices to evaluate the research landscape. The science mapping tactics made use of the VOSviewer and Biblioshiny software.

The performance and science mapping analysis shows that recent research on the digital divide has not been sufficiently exposed and examined. The analysis discovered emerging topics, prolific authors and nations, affiliations, a network of collaboration among authors, countries and institutions, bibliographic coupling and keyword co-occurrence.

This work presents a state-of-the-art that has significant theoretical and practical ramifications for the existing digital divide literature. The methodologies and database used in the current study are more extensive.

This study aims to examine the research output on digital divide from 2001 to 2020 and measure the qualitative and quantitative growth of literature during the stated period by using required bibliometric measures for identifying the types of documents, yearly growth, country productivity, citation network of collaborative countries, authorship pattern, top authors, cocitation networks and assorted facets.

Web of Science database was used to retrieve the required data for this study. Keeping the objectives of this study in mind, the keyword “Digital Divide” was used as the search term. Moreover, the retrieved data were limited from the year 2001 to 2020 for two decades. A total of 5,518 publications were filtered and focused for subsequent facet-wise analysis and interpretation. Required bibliometric indicators like types of documents, yearly growth, authorship pattern, degree of collaboration (DC), country productivity, h-index and citation impact were used to study various dimensions of publication trends. VOSviewer software was used to visualize the authorship network, bibliographic coupling and keyword occurrences.

This study finds a total of 5,518 publications on the topic digital divide contributed by 14,277 authors from 130 countries across the world published through 2,843 source titles in 13 global languages during the past two decades (2001–2020). The annual growth of publications (AGP) on the topic digital divide shows 38.43% AGP globally. Journal articles have been identified as the preferred type of document with 73.11% of the literature. The DC indicates a healthy trend of collaborative research with a mean value of 0.70. The USA is the table topper with the contribution of 1,933(35.03%) publications and 77 h-index and James J., from Tilburg University, The Netherlands, is identified as top amongst the most productive authors with the highest number of 34 publications (h-index 14).

This study restricts its scope on research productivity to the theme “digital divide” regarding authorship pattern, DC, most productive authors, most productive countries, most published sources and other key facets. This study exclusively refers to the Web of Science database in retrieving the required data. Moreover, this study takes global research into account with no geographical or language limitations and comprehends literature on digital divide for two decades ranging from the years 2001 to 2020.

Teachers and research scholars interested in bibliometric studies can benefit from insights into the scholarly documents published on the topic digital divide from 2001 to 2020.

This study yields some interesting findings on published literature on the digital divide during the past two decades relating to the most striking contributions, highly cited journals, the most prolific authors, country productivity, keyword cooccurrence and assorted parameters.