Deep Learning-enabled Smart Process Planning in Cyber-Physical System-based Manufacturing.

AuthorValaskova, Katarina
  1. Introduction

    The convenience of a massive quantity of data has facilitated the vast use of machine learning and deep learning approaches throughout various industrial sectors entailing computer-based critical systems. (Bernardi et al., 2020) Machine learning algorithms are instrumental in planning and decision making in smart manufacturing. (Chang et al, 2020)

  2. Conceptual Framework and Literature Review

    Machine learning, deep learning, and big data computing further the advancement of smart supply chains. (Wan and Qie, 2020) Industry 4.0 pursues the setting up of self-governing and dynamic operations (Kenrick et al, 2019; Lafferty, 2019; Mircica, 2019; Popescu et al, 2019; Valaskova et al, 2018) to set off the large-scale manufacturing of highly made-to-order goods. (Asif, 2020) Industrial Internet of Things technologies enable the collection of huge volumes of data (Kral et al., 2019; Lazaroiu et al, 2020; Popescu et al, 2017; Trettin et al, 2019) that are harnessed to train and use artificial intelligence algorithms (Kliestik et al., 2018; Lazaroiu, 2018; Nica et al., 2014; Slaby, 2019; Wingard, 2019) to decide on intricate industrial issues, precisely and automatically. (Ortego et al, 2020)

  3. Methodology and Empirical Analysis

    Using and replicating data from Deloitte, KSM, PwC, SME, Statista, and Tractica, we performed analyses and made estimates regarding top challenges to implementing smart manufacturing solutions (%) and business organizations' reasons for adopting artificial intelligence (%). Data were analyzed using structural equation modeling.

  4. Results and Discussion

    Production systems are decisive in smart manufacturing as a significant amount of manufacturing resources are fashioned and advanced with virtual/digital ones related to the physical ones across their growth. (Zhang et al., 2020) The rise of the Internet of Things has bolstered the progress of data-driven cyber-physical systems that gather and mine data that is spatially and temporally complex. (Karunarathne et al, 2020) (Tables 1-9)

  5. Conclusions and Implications

    Industrial Internet of Things gateways handle heterogeneous wired and wireless networking technologies by harnessing various protocols, in addition to performing cutting-edge data mining (e.g., machine learning algorithms and big data analytics) and operational supervision of the system by using interconnected sensors and actuators that are Industrial Internet of Things devices that specifically assess system variables (e.g., temperature, vibrations, motion, etc.) and accomplish certain actions on it. (Corallo et al., 2020)

    Note

    The interviews were conducted online and data were weighted by five variables (age, race/ethnicity, gender, education, and geographic region) using the Census Bureau's American Community Survey to reflect reliably and accurately the demographic composition of the United States. The precision of the online polls...

To continue reading

Request your trial