The Automation Stack

Automation covers a LOT of ground. When someone is talking about automation they might be talking about that small machine over there on the plant floor, a large facility with thousands of sensors and outputs, a SCADA system with thousands of sites, or a software automation and visualization.

It can be helpful to think of and describe parts of automation as layers in a stack. This stack ranges from the specific sensors and processes on the plant floor (Layer 0) all the way up to the business enterprise resource planning or cloud data analytics (Layer 4/5).

In this article we are going to do a quick review of the Automation Stack, as taught by Industry 4.0 Solutions at iiot.university. Also taking inspiration from the ISA-95 Automation Stack.

The Automation Stack

The Industry 4.0 Automation Stack

Typical Data Flow

Industry 3.0 Traditional Thinking

In a Industry 3.0 company each layer in the stack is somewhat isolated and data must flow up through each layer of the stack. That means if ERP needs sensor data it needs to flow and be mapped from Layer 0 all the way up to Layer 5. As you can imagine this incurs a significant programming and engineering effort at each layer of the stack. What that means in practice is only small bits of data from the plant floor end up being available at the higher layers in the stack.

Industries that still follow this traditional, legacy practice today are mining, oil and gas, energy, and pharmaceuticals. Moving data this way is still incredibly common and is promoted through cyber security network models like Purdue.

Industry 4.0 Paradigm

Luckily there is a new way of thinking, one that does away with all that manual mapping. Industry 4.0 companies have all the same layers of the stack but they deploy a digital infrastructure that makes it possible for data to flow directly and securely from one layer of the stack to another. Data can and should still flow bottom up (i.e. Field -> PLC -> HMI for local process control) but systems like ERP or MES can get information quickly and easily from lower layers directly from the Digital Infrastructure. Interestingly this also allows edge and process control systems to consume data from higher layers which creates new possibilities and possible solutions for plant floor teams.

This new stack is kept secure with encryption certificates, logging, authentication, and usually a zero trust network model. Purdue can work here but it needs to be broken to allow the new data flows to work properly.

Note: One way of architecting this digital infrastructure is by building a Unified Namespace (UNS) on an open transport protocol like MQTT. Your digital infrastructure does not have to be MQTT, but this way of building a UNS is easy to implement, scales well, and is very open.

The Digitally Transformed Automation Stack

Where does JPI fit?

JPI Solutions is an Industry 4.0 integrator and we have experience working in and integrating data at all the layers of the stack. Our bread and butter is Layers 0 thru 3 though with heavy focus on Layers 1 and 2. If its something that needs controlling or needs to get connected, we would love to partner with you to make it happen.

Keep following us for more Industry 4.0 information.

Thanks and credit to Industry 4.0 Solutions and their training found at iiot.university.