Driving Industry 4.0 at Distributed Edges with Cloud Orchestration
time , an architecture for Industry 4.0 must be able to support dynamic , data-driven production optimization .
Edge computing is a promising approach to mitigate risk and to avoid that production or assembly lines do not stop due to intermittent connection to the cloud . Typically , edge computing is performed in close proximity to the factory shop floor or warehouse , leveraging connectivity to both enterprise business solutions and onsite Manufacturing Execution Systems ( MES ), Programmable Logic Controllers ( PLC ), sensors , and other industrial automation systems . Consequently , this approach helps to achieve the targeted Key Performance Indicators ( KPI ) such as Overall Equipment Effectiveness ( OEE ). For these reasons , edge computing is an essential element in the reference architecture for Industry 4.0 [ 2 ]. However , it is crucial to develop strategies addressing the higher complexity in edge computing approaches and keeping the costs ( e . g ., development , integration , and operational ) under control .
This paper introduces the approach SAP recommends to enable faster decision making on premises and to achieve continuity of business-critical processes in manufacturing by extending a subset of cloud capabilities to the edge , close to the source of data . Challenges related to deploying , managing , orchestrating , and operating distributed business applications in different manufacturing sites are considered and addressed to guarantee the targeted business outcome . This paper also discusses the approach and gives recommendations where to deploy business applications based on latency and criticality . Based on scenarios for Industry 4.0 introduced in [ 3 ], selected mission critical scenarios for logistics and manufacturing will be discussed to elaborate the “ added value proposition ” of the solutions presented in this paper . Furthermore , we give an overview to the possible changes and improvements on the shop floor and in the supply chain .
EDGE COMPUTING – CHARACTERISTICS , CHANCES AND CHALLENGES
Table 1 introduces the dimensions characterizing the edge computing and cloud computing paradigms . Their comparison emphasizes the chances and challenges to be addressed by applying edge computing approaches to manufacturing processes that will be discussed in this section .
Table 1 - Dimensions of edge computing and cloud computing . |
Dimension |
Edge |
Cloud |
Latency : |
low |
high |
Number of nodes : |
many |
few |
Computing and storage resources : |
from constrained lightweight edge devices to heavy edge nodes with more resources |
nodes with high to very high computing and storage resource ( hyper-scaler ) |
Topology : |
distributed |
centralized |
Complexity : |
heterogeneous ( devices , industry protocols ) |
homogenous |
- 20 - June 2021