The industrial IoT and AI’s role in it stem from Accenture’s incorporation of the IIoT as it moves over to AIP+. Accenture states that AIP+ is not just regular AI, but rather a combination of AI with analytics, automation, and data, all in one neat package. Accenture’s development of the IIoT has centered mostly on creating a system that emphasizes the balance of edge computing with cloud computing, with more focus placed on edge.
Accenture’s edge computing framework builds upon existing technology developed on the open-source Eclipse Foundation’s Kura. Clients of Accenture now have the freedom to shift processing logic to edge instead of keeping it on the devices, allowing for the development of different IIoT edge gateways. The innovation was the basis of Accenture’s initial launch as a CPaaS. Today, however, Accenture operates out of Microsoft Azure or Amazon AWS.
Automation of IIoT is ideal for use with SAP S/4 HANA. Devices running at the edge can fire alerts to the SAP system which can then create alerts of scheduled maintenance or replacement of parts of devices as the need arises. The downside of gateways through edge servers means that there is usually edge middleware required to collect data from the IoT devices on the ground. Gateways allow for flexible application of computing power, although for businesses running older hardware, limitations may occur because of the age of their system components.
AI and Analytics
As noted before, AIP+ utilizes AI in a unique way that Accenture terms “Applied Intelligence.” The applied intelligence system uses the Accenture Insight Platform as a basis for deep learning and machine learning. At its latest iteration, AIP allows users of Accenture’s software to deploy preconfigured solutions for machine learning as well as develop their own without requiring the services of a data scientist.
AIP’s power comes from the use of open-source technology, along with application programming interfaces (APIs) and use of utilities that allow for edge computing and machine learning on the edge. Pre-built solutions offered to clients cover a wide range of options and include useful add-ons like fleet management with optional route optimization and remote monitoring of processes combined with potential failure prediction, as used by wagering sportsbooks like top NFL betting sites.
The Next Iteration of AIP
AIP+ is supposed to be an upgraded and expanded AIP system incorporating new, more efficient methods of interacting with the IIoT. One of the fundamental changes will be the replacement of Eclipse with EdgeX Foundry, which is supported by the Linux Foundation. The implication is that users will have access to a broader pool of talent and open-source software at their disposal. AIP+ also intends to have better, higher-frequency time-series processing at edge nodes.
What Are The Impacts on SAP Customers?
Accenture’s shift to AIP+ offers a lot of benefits to SAP customers. However, Accenture advises SAP customers that they should do the following to ensure compatibility with the new AIP+ system:
- Figure Out Which Machines Need Connecting: Metrics are only as good as the data provided. It makes no sense connecting machines that don’t provide actionable data. IT & OT professionals need to come to a consensus as to which data points would be most beneficial to improving the company’s efficiency.
- Use a Test Case for Business Applications: Edge gateways, while a useful resource, isn’t a magic bandage that cures all ills. However, they can significantly impact the processing and compute power of a company’s IIoT Applications. Using a test case may help businesses gauge whether edge gateways would be better for their deployment.
- Balance Edge and Cloud Application: Cloud computing has its uses in training AI quickly and processing vast amounts of data cheaply. Companies need to see where they are better off using cloud computing and where wedge computing provides more benefits.
- Prioritize AI Capabilities: Continual iteration leads to better trained AI systems. Companies need to determine how to implement IIoT training so that businesses can deploy machine learning models to the edge.
The aim is to provide companies with actionable advice that can enhance their adoption of IIoT and AI in a significant way. The final decision rests with the company, but the framework allows for an impressive depth of deployment for companies.