https://de.wikipedia.org/wiki/ISA-95

ANSI/ISA-95 ist eine Norm für die Integration von Unternehmens- und Betriebsleitebene, die von der ISA herausgegeben wurde

There are 5 parts of the ISA-95 standard. Dont confuse them with “levels”

• Teil 1 Models and Terminology[1]
• Teil 2 Object Model Attributes[2]
• Teil 3 Models of Manufacturing Operations Management[3]
• Teil 4 Object models and attributes of manufacturing operations management[4]
• Teil 5 Business-to-Manufacturing Transactions[5]

https://www.plm.automation.siemens.com/global/en/our-story/glossary/isa-95-framework-and-layers/53244

ISA-95 incorporates the layers model of technology and business process for manufacturing enterprises as levels for the standard. These levels are:

• Level 4
  • Defines the business-related activities needed to manage a manufacturing operation.
• Level 3
  • Defines the activities of workflow to produce the desired end products.
• Level 2
  • Defines the activities of monitoring and controlling the physical processes.
• Level 1
  • Defines the activities involved in sensing and manipulating the physical processes.
• Level 0
  • Defines the actual physical processes.

From a component or software perspective, Levels 1 to 4 can be seen like this:

• Level 4: Business logistics systems (e.g., ERP)
• Level 3: Manufacturing operations systems (e.g., MES)
• Level 2: Control systems (e.g., PLCs, DCSs)
• Level 1: Intelligent devices

See https://www.brightlysoftware.com/blog/introduction-iot-cmms-and-isa-95-framework

Information Technology vs Operational Technology

In IIOT context what is Management Execution Systems (MES)?

Comment:

• In my experience, customers should be told that MES is not an IT system, but an Enterprise layer!
• The functions in it can even be autonomous IT solutions as modules in the sense of Industrie 4.0!

Manufacturing Execution System

https://docplayer.org/23562722-Mes-manufacturing-execution-system.html

• Digitalization uses the foundational Industry 4.0 technologies such as the Asset Administration shell to create a Digital Factory. I am deliberately using the term “foundational Industry 4.0 technologies” here as the general Industry 4.0 definition is much wider, including communication and automation technologies.
• Networking uses communication technologies in addition to the foundational I4.0 technologies and thus enables the Industrial Internet of Things.
• Cloud technologies featuring MES, Advanced Analytics, etc. enable Advanced Automation which characterizes the Smart Factory. Advanced Automation orchestrates input from multiple sources and runs real-time operations as well as added value services.
• Finally, AI enables some adaptation and self-optimization of the factory. I have added this additional step essentially to cover the Autonomous Factory which is also referred to as “Dark Factory” as no light is required here due to the absence of workers.

Comment:

• i‘d guess iiot comes before, or better, in-parallel to digital factory. For example, mqtt is fully semantic-less. Connectivity is also seen as a pre-step in industry 4 maturity model. https://opcconnect.opcfoundation.org/wp-content/uploads/2018/02/6Steps-800x452.png
• „cloud technologies“ should be called differently maybe „management and optimization services“ like Namur NOA does it. These services are arguably not focusing on the deployment as you indicated by mixing edge and cloud.

=== From https://www.mckinsey.com/capabilities/mckinsey-digital/our-insights/how-to-build-a-data-architecture-to-drive-innovation-today-and-tomorrow

One pharmaceutical company, for example, is setting up an internal “data marketplace” for all employees via APIs to simplify and standardize access to core data assets rather than relying on proprietary interfaces. The company is gradually—over 18 months—migrating its most valuable existing data feeds to an API-based structure and deploying an API management platform to expose the APIs to users.

Field devices can talk directly to the cloud using “OPC UA”.

OPC UA offers an option to model your device model.

This is happening via a In-Memory model

See