Machine Malfunction Chatbot
A chatbot solution to support manufacturers by reducing the downtime of automated machines and thus improving the stability of manufacturing processes.
Our Final Result
Overview
Challenge
The main problem that manufacturers faced with their automated machines was downtime due to malfunctions. The longer it takes for a malfunction to be noticed and addressed, the longer the machine goes without producing anything. The current solution required workers to have visual contact with the machine to determine whether it had to be stopped. Without this visual contact, a malfunction can go unnoticed for an extended period of time. Therefore, it is always necessary for someone to be nearby to monitor the machine’s progress, 24/7.
Solution
We developed a solution that consisted of a chatbot implemented on a single board computer which monitors the status of the machine as it is working. Whenever the machine stops due to a malfunction, the controller collects information about the malfunction, including images and error messages, and sends this information to a group chat of workers responsible for the machine. A final decision as how to proceed is sent to the chatbot. The controller then instructs the machine to either continue working or wait for further assistance. Our chatbot solution helps manufacturers reduce the downtime of their machines, thus improving the stability of their manufacturing process.
Technologies
- Sakis 3G
- Python 3
- PiXtend V2-S-
- fswebcam
- Raspberry Pi
Impact
Decreased downtime in manufacturing processes. Through our chatbot solution, instead of waiting for someone to notice that production has stopped, the machine will notify those responsible for it that it needs assistance. The machine acts like a human by sending this notification to a group chat and therefore giving the user the impression of chatting with a real person.