industrIA

• 

  industrIA

• 

  Prototype industrIA

Inspiration

Coming from an engineering background and having carried out internships in several companies in different sectors (pharmaceutical, food processing, hydraulics), we have noted an inadequacy in the organization of maintenance, particularly in the monitoring of the various machines as well as the process of repairing them. But also, a difficulty in accessing the necessary information to solve the problems encountered, which leads to a loss of time and income, a negative impact on the image of the company and consequently a financial decline and even the collapse of a company.

What it does

IndustrIA is an intelligent system based on high-performance artificial intelligence models that allows for better organization of maintenance within an industry through a cross-platform software with several interfaces:

• a main interface gathering the different performance indicators, called KPI, allowing to maintain a global view on all the machines and to guarantee a regular follow-up to ensure a precise and effective intervention.
• an interface allowing users to recognize the machine and to visualize its details through a 3D model (that we generate based on the machine and its documentation) to access the information and the maintenance history of each main component of it.

How we built it

We started by designing the interfaces using the Figma tool. Then, we used the well known software Solidworks and Blender to create the 3D models of the machines. For the development of the artificial intelligence and vision models, we used Keras, Tensorflow and openCV with the programming language Python. Finally, for the creation of a demo version of the interfaces, we opted for a web development where we used for the frontend part the HTML, CSS and javascript languages and for the backend part, the Flask framework.

Challenges we ran into

During the realization of our project, we faced several challenges including the optimization of artificial intelligence models, 3D modeling which is usually very complicated to do but still we were able to obtain very good results.

Accomplishments that we're proud of

We were able to realize a demo version of our solution and apply it on a laptop. We started by modeling it in 3D and then extracting the different information about it and visualizing it through a web platform. We also got very good results in the artificial intelligence models we developed.

What we learned

We learned lot by participating in this hackathon and devloping our solution from scrach especially in teamwork.

What's next for industrIA

As a perspective, we aim in the near future to launch industrIA as a startup that offers this service for companies. We also aim to develop our artificial intelligence models through the enrichment of our database. Preventive or predictive maintenance is always present in our vision of the future as it fits perfectly in our objectives.

Built With

• blender
• css
• html
• javascript
• keras
• opencv
• python
• solidworks
• tensorflow