A manufacturing system can be fully autonomous if it can fully understand its data and its surrounding environment and execute decisions with less human intervention.

Data in the manufacturing context can be broadly classified into 4 types – structured data (a sequence of data or time series information from sensors); unstructured data like images from cameras used by robots (automated robotic systems) for navigation or automated product classification; a combination of structured and unstructured data, like in case of augmented reality where you superimpose images and text to derive an outcome; and fourth being behavioural and surrounding data where devices learn from interactions and integration, the cognitive aspects of the devices.

Using AI technologies helps us to derive intelligent insights for each of the above data scenarios, For instance, machine learning models can be built for analysing the time series information from sensors for anomaly detection, condition or predictive based maintenance and output can be fed to other systems to take corrective action, like ordering new parts.

Deep learning models like convolutional neural networks can be used for computer vision, which can be used by robots for automating any activity involving image analysis, like production categorisation, product placement, navigation, etc. Optimisation models can be employed at every step of process to identify bottlenecks and improve efficiency. I also feel, there will always be some tasks that cannot be fully automated and humans and robots will work together to achieve the highest level of automation.

For more details, refer to the use case – Application Of IoT In Manufacturing