Modern automotive manufacturing facilities generate data from PLCs, robots, RFID readers, RTLS devices, MES platforms, machine vision, AGVs, and production stations.
The challenge is not in collecting this data, but rather, in transforming it into meaningful operational context.
This is where AIoT (Artificial Intelligence of Things) comes to play.
From Automotive Data to Operational Context
Let’s take a single vehicle that is being assembled in a factory.
A RTLS system may know its physical location while RFID may identify the part or asset.
The MES may know what production stage the vehicle is in while machine or PLC data may indicate what is happening at this workstation.
Each system may provide a piece of the puzzle, but only by combining these data sets into the single operational context we can truly understand what is happening on the factory floor.
A typical data architecture may leverage OPC UA for connectivity, MQTT for lightweight event and telemetry messaging, RFID and UWB RTLS for seeing where assets are, BLE for connected devices and worker telemetry, Edge for smart data processing and MES/ERP integration to bring production and business events into this context.
The ultimate goal is not just to connect more devices, but rather, is to enable new levels of context-aware operations.
Understanding Context, Not Just Collecting Data
For instance, if we know about a vehicle’s VIN, the location data may enable us to understand where is this particular vehicle or part heading to.
But, when combined with production data, this information can help us answer more interesting questions such as: is the asset where it is supposed to be at this stage of the process?
Do we have the right tools and materials ready for it?
What production stage are we at?
What events have occurred?
Which systems need to be notified of the change?
Here we can see how various technologies such as RTLS, RFID, industrial telemetry, and manufacturing software, can complement each other rather than existing in complete isolation.
Architecting the Solution
For the developers who work on industrial systems, the question is of not just collecting more data, but really, about delivering more context.
This means making decisions about the data architecture, event processing, edge technologies, system integration and interoperability, industry-specific data models and many other considerations.
AIoT can enable the needed level of contextual intelligence, but it is up to developers to ensure that devices, data, applications, and operational workflows form a coherent whole.
In automotive smart factories of the future, the emphasis will be not on simply making more machines connected, but rather, on making sure the data that is being collected is actually useful for operational analysts, managers, and workers.
Top comments (0)