No abstract

Deep learning techniques, especially Convolutional Neural Networks (CNN), dominate the benchmarks for most computer vision tasks. These state-of-the-art results are typically obtained through supervised learning, for which large annotated datasets are required. However, acquiring such datasets for manufacturing applications remains a challenging proposition due to the time and costs involved in their collection. To overcome this disadvantage, a novel framework is proposed for data augmentation by creating synthetic images using Generative Adversarial Networks (GANs). The generator synthesizes new surface defect images from random noise which is trained over time to get realistic fakes. These synthetic images can be used further for training of classification algorithms. Three GAN architectures are trained, and the entire data augmentation pipeline is implemented for the Northeastern University (China) Classification (NEU-CLS) dataset for hot-rolled steel strips from NEU Surface Defect Database. The classification accuracy of a simple CNN architecture is measured on synthetic augmented data and further it is compared with similar state-of-the-arts. It is observed that the proposed GANs-based augmentation scheme significantly improves the performance of CNN for classification of surface defects. The classically augmented CNN yields sensitivity and specificity of 90.28% and 98.06% respectively. In contrast, the synthetically augmented CNN yields better results, with sensitivity and specificity of 95.33% and 99.16% respectively. Also, the use of GANs is demonstrated to disentangle the representation space and to add additional domain knowledge through synthetic augmentation that can be difficult to replicate through classic augmentation. The proposed framework demonstrates high generalization capability. It may be applied to other supervised surface inspection tasks, and thus facilitate the development of advanced vision-based inspection instruments for manufacturing applications.

The focus of this article is to develop a semi-Markov processbased analytical approach that is faster and more accurate than the simulation approaches used in the existing software programs such as RAPTOR (ARNIC, Annapolis, MD), BlockSim (Reliasoft, Tucson, AZ), etc., for system availability analysis. The steady-state solution of the semi-Markov process (SMP) model provides system availability. For mechanical systems and their components, the degradation rate increases with the aging process. A Weibull distribution for the time to failure is appropriate for such systems. The semi-Markov model does capture dependencies of the repairable system. A steady-state solution was obtained by a two-stage analytical approach, which was validated by Monte Carlo simulation.

PurposeIndustry 4.0 and circular economy are the two major areas in the current manufacturing industry. However, the adoption and implementation of Industry 4.0 and circular economy worldwide are still in the nascent stage of development. To address this gap, the purpose of this article is to conduct a systematic literature review on integrating Industry 4.0 and circular economy. Further, identify the research gaps and provide the future scope of work in this area.Design/methodology/approachContent-based analysis was adopted for reviewing the research articles and proposed a transition framework that comprises of four categories, namely, (1) Transition from Industry 3.0 to Industry 4.0 and integration with circular economy; (2) Adoption of combined factors and different issues; (3) Implementation possibilities such as front-end technologies, integration capabilities and redesigning strategies; (4) Current challenges. The proposed study reviewed a total of 204 articles published from 2000 to 2020 based on these categories.FindingsThe article presents a systematic literature review of the last two decades that integrates Industry 4.0 and circular economy concepts. Findings revealed that very few studies considered the adoption and implementation issues of Industry 4.0 and circular economy. Moreover, it was found that Industry 4.0 technologies including digitalization, real-time monitoring and decision-making capabilities played a significant role in circular economy implementation. The major elements are discussed through the analysis of the transition and integration framework. The study further revealed that a limited number of developing countries like India have taken preliminary initiatives toward Industry 4.0 and circular economy implementation.Research limitations/implicationsThe study proposes a transition and integration framework that identifies adoption and implementation issues and challenges. This framework will help researchers and practitioners in implementation of Industry 4.0 and circular economy.Originality/valueReviews of articles indicated that there are very few studies on integrating Industry 4.0 and circular economy. Moreover, there are very few articles addressing adoption and implementation issues such as legal, ethical, operational and demographic issues, which may be used to monitor the organization's performance and productivity.

Type 2 diabetes mellitus (T2DM) is one of the most threatening, non-communicable ailments worldwide.