Masterstudent presenterade examensprojekt på forskningskonferens
Joel Göransson, masterstudent på civilingenjör i datateknik presenterade sitt examensprojekt "AI-Driven Predictive Maintenance for Industrial Automation" på forskningskonferensen SNCNW - Swedish National Computer Networking and Cloud Computing Workshop.
Läs hela intervjun på engelska nedan:
Joel Göransson's paper, "AI-Driven Predictive Maintenance for Industrial Automation," focuses on the cutting-edge field of Industry 4.0. This research addresses the crucial need for predictive maintenance in industrial settings to minimize downtime and maximize productivity.
Can you give us a very brief overview of your paper?
In the submitted conference paper, it was explored how predicting maintenance can be done within the manufacturing industry. In particular, it was explored how various models can be created using AI to determine if the maintenance is required or not. A variety of algorithms were looked at and how different parameters affect the performance of these models.
Were there any unexpected or particularly interesting findings or challenges that you encountered during the research process?
During the research process I got to learn a lot of interesting things about creating AI-models, such as how data can be transformed to yield better accuracy and which algorithms works better. However, the most interesting thing for me was looking into how Explainable AI can be used. This technology aims at explaining why a prediction is made within AI-models. This can for instance be useful within predictive maintenance, as it can help explain why a machine needs maintenance, and which parameters that affect the decision the most.
Could you tell us a little bit about the experience of traveling to and attending the conference? And of course, tell us about the experience of presenting your paper!
To travel to Kristianstad for the conference will be a life-long memory for me. I got to meet a lot of brilliant people that were truly passionate about their research. It was really fun to present my own research at the conference, with the added bonus of getting to answer some great questions and having some excellent discussions. In addition to this, I also got to explore Kristianstad, including visiting a film museum, and having some great food.
How do you see the field of industrial automation and predictive maintenance evolving in the future and what role do you envision AI playing in it?
Predicting when maintenance needs to occur, rather than having regularly scheduled maintenance, will become more and more popular in the future. This will allow companies to save money as equipment can last longer and downtime can be reduced for the machines with knowledge in advance. AI will be a big part in making these types of predictions, as well as making the manufacturing process more efficient and flexible.
Joel discussing his paper with an SNCNW attendee.
Read more about the Swedish National Computer Networking and Cloud Computing Workshop through the link below.