Application deployment framework for large-scale Fog Computing environment (original) (raw)

2020, Coventry University

The concepts of Industry 4.0 provide a new means of integrating concepts from ubiquitous computing with manufacturing technologies through cybernetics. This advances the automation of the manufacturing systems and helps improve product quality, production efficiency, condition monitoring and decision making (J. Lee, Bagheri, and Kao 2015; DIN 2016). Within this concept, machines become connected with humans through computer systems to work in a coordinated way to automate data acquisition, sharing and exchange among the physical and virtual worlds. The wide spread availability and affordability of sensors, wireless networks and the accessibility of high-speed Internet make real-time multiple parameters monitoring and control of manufacturing process possible in a way that was not feasible before (Y. Lu 2017). This leads to a great number of sensors being deployed to physical machines which in turn generates a large volume of data that requires computationally intensive analysis and interpretation for decision-making purposes. The resulting decisions, whether made by humans or software, often need to be transformed into control signals for actuators to operate the machine in the physical world. This then creates a loop-back to the sensor system as new sets of data are collected and sent back for further analysis, reflecting changing machine states over time. This type of system based on Cyber-Physical System (CPS) is a facilitator for realising the concepts of Industry 4.0. It enables computational algorithms and physical components to interact with each other through real-time monitoring and control to improve productivity (Trappey et al. 2016; L. Wang, Törngren, and Onori 2015). Yet, as stated in (Wiesner, Marilungo, and Thoben 2017) traditional servers with limited capacities may not be able to cope with the new challenges in terms of scalability and complexity of such systems. In turn, 1.3 Research Aims and Objectives The research aim can be defined as the broad challenge of this work and is stated below. Formulate, implement and evaluate an IoT and Fog based Application Deployment Framework for Industry 4.0 systems The objectives can be described as steps that need to be taken to fully answer both the main research question and the more precise questions that make it up. Due to the nature of the work the objectives can be split up into three categories, Platform, Model and Method. Each of these looks at distinct components of the big framework. 1.4.2 Fog of Things Platform The first component was designed to answer the requirements of Industry 4.0 and also to explore some of the novel concepts of IoT and Fog Computing. To decide which components are to be used on the platform a similar methodology was used as in (Cruz et al. 2018) where