THE CLIENT
The company is a leading player in the motorcycle manufacturing sector, constantly striving to maintain its leadership by improving its products and production processes.
NEEDS
The company expressed the need to implement an automatic defect recognition system for motorcycle wiring defects in the fuel, braking, and electrical systems based on photographs taken directly on the production line.
THE SOLUTION
The automatic wiring defect detection system was created using ADR-Flow, our system developed specifically for this purpose. It is divided into two modules: the edge module and the cloud module. The edge module, the part of the system in direct contact with the production line, was configured with:
- a control station, which manages interaction with the operator via a graphical and audio user interface and a barcode scanner that allows vehicle identification; this also organizes the sending of photos and metadata to the remote server (cloud module);
- three “smart” cameras, which are real calculators (Raspberry in this case) with powerful built-in functions that allow for automatic recognition of defects in real time
The cloud module, on the other hand, consists of a virtual machine that has the task of:
- receive, record and analyze images
- allow specialized technicians to search for images, view them and manually classify them, i.e. say whether they represent a defect or not, in order to use them for training and continuously verifying the performance of neural networks;
- support data scientists in training neural networks
- provide the system manager with a dashboard to monitor the system through simple graphs and tables with the distribution of defects over time, the image acquisition sequence on a daily, weekly or monthly scale, the analysis of the line speed over time and the quality of the recognitions performed;
Once trained, the neural network is sent to the camera to perform real-time recognition.
RESULTS
The company’s needs for industrial productivity have been met, achieving great results and benefits, such as:
- improving output quality;
- greater efficiency and therefore lower costs of the production process;
- possibility of activating continuous improvement processes in production, thanks to the learning capacity brought by these techniques;
- possibility of carrying out in-depth analyses on the causes of defects with the consequent opportunity to better engineer the process itself.