During the 4th to 6th of April the Advanced Factories Congress and Expo was held in Barcelona (Spain). In this congress, some of the latest technology and approaches for factory automation were on display. This new generation of industrial development is also known as Industry 4.0, which has become a buzzword for the next industrial revolution in which Cyber-physical systems, 3D-printing, Advanced Analytics, Artificial Intelligence and the Internet of Things, among others, are being incorporated into the manufacturing environment.
IRIS‘s Marketing Manager, Carlota Feliu gave a presentation (see above photo) of ‘Hypera’ , IRIS‘s in-line foreign body sensor for the food processing industry. Dr. David Nettleton and Alexandr Tretyak, of IRIS‘s Data Science team also attended, visiting the Expo and selected conference sessions.
Some key take-aways from the congress/expo were:
- Collaborative Robots: industrial robots which can work/move in the same space as humans (Eurecat, see photo).
- 3-D Printing / Additive Manufacturing
- Artificial Intelligence + Analytics
- Complex Simulation Models (digital twins or Doppelgänger) of the physical systems
- Internet of Things (IoT): In-line big data capture from sensors
- Cloud computing
Now let’s take a slightly ‘deeper dive’ on three of these themes: AI Analytics, 3D Printing and Collaborative robots
Artificial Intelligence + Analytics
AI Analytics can provide the means for simulation in every stage of the life cycle of a product (from model and design, to functionality prediction). This is especially useful for launching new products in already existing plants by building simulation models which are used to evaluate and verifying the impact on production and human-machine interactions. Once deployed, Big Data analytics tools can be used to analyse sensor data uploaded to the cloud from the factory floor. The capacity to support and control big flows of information is one of the most important applications of Industry 4.0.
Data fusion techniques can give a ‘double whammy’ by simultaneously selecting data based on (A) relevance and (B) reliability for different ‘dashboard’ indicators. Machine learning algorithms can be used to train complex multi-variate models with in-line machine parameters and telemetry data. The trained models can then be executed in real time to give calibration or alert support, and can be retrained dynamically if their predictive precision falls below a given threshold.
3D Printing, also known as additive manufacturing (AM), refers to processes used to create a three-dimensional object in which layers of material are formed under computer control to create a physical object. Objects can be of almost any shape or geometry and are produced using digital model data from a 3D model or another electronic data source such as an Additive Manufacturing File (AMF) file.
Two main technologies used are:
- Fused deposition modellng (FDM) in which the model or part is produced by extruding small beads or streams of material which harden immediately to form layers.
- Photopolymerization / stereolithography to produce a solid part from a liquid. Inkjet printer systems spray photopolymer materials onto a build tray in ultra-thin layers (e.g. between 16 and 30 µm) until the part is completed.
In the photo we see some examples of surgical implants and prosthetic devices generated by Hewlett Packard’s 3D printers.
Collaborative and Smarter (or less dumb) Robots
AI will facilitate greater autonomy for robots. For example, in a previous situation a robot would repetitively fix one type of screw in orifices with a given diameter. In a new scenario, a robot will be able to autonomously choose from several types of screw depending on the orifice diameter. Also, AI can help to improve the ability of robots to work together with humans. For example, in the implementation of safety protocols for operators working in the same area by, using touch or vision sensors can detect when a human enters in a risk area in close proximity to the robot, causing the robot to stop or take evasive action.
A “collaborative” robot is a robot intended to physically interact with humans in a shared workspace. This is in contrast with other robots, designed to operate autonomously or with limited guidance, which is what most industrial robots were up until the decade of the 2010s. In the photo we see the author interacting with a humanoid robot (PAL Robotics) at the Expo.
One challenge is being able to integrate all of these aspects to facilitate, for example “mass production of customized products“. Although the individual technologies are already available, it will probably be several years before we see them completely integrated in real industrial environments, together with the value chain extending from the factory floor to the final customer.
IRIS is working on several Industry 4.0 projects, such as Openmind (in-line machine learning calibration applied to mass production of customized medical devices) and CTC+ (embedded machine learning algorithms for modelling grain silo loading and predictive indicators for grain quality monitoring), in which we are taking these challenges to the next level in collaboration with leading edge industrial and technology partners throughout Europe.