Habber Tec observed that in some companies there is a strong interest in investing in quality and sustainability, mainly those companies in an expansion phase or with strong national and international competition, especially in Europe, which demands very high quality standards.
The review of each of the products manufactured, one by one, is a cost that is totally unaffordable for any company, so in some cases complex and expensive visual recognition systems linked to highly complex deep learning algorithms are used to detect faults and defects in manufacturing. This does not guarantee the detection of problems either, as this type of solution only checks the external appearance of the product.
This is why DigiQual, a project supported by CDTI innovation and the European Regional Development Fund (ERDF), was created, which tries to provide an answer to this problem thanks to the information obtained from the assembly lines, this information is processed by a machine learning model that has been previously trained with all the history of the previous assembly line.
Objectives and resolution
The DigiQual project aims to go beyond traditional supply chain approaches, with the use of artificial intelligence (AI), internet of things (IoT) and Big Data to achieve the new quality paradigm of the Industry 4.0 era of Cyber-Physical Production Systems (CPPS). DigiQual’s approach aims at reducing quality testing in the production environments of factory assembly lines, for which it will be integrated into the production chain with feedback loops and feeding of the generated information.
Thus, the main objective is to create an automatic system capable of predicting, based on the manufacturing conditions of a product, the probability that it will not need to perform a quality test based on the paradigm of artificial intelligence (AI), internet of things (IoT) and Big Data.
This project aims to cover the following specific technical objectives:
- Creation of a data lake where all the information received from the assembly line will be stored.
- Creation of an integration system with the assembly line that is responsible for receiving or obtaining all the information and storing it in the data lake.
- Creation of an ODS database that stores in a relational model all the information obtained from the assembly lines.
- Creation of an intelligent prediction system with artificial intelligence capable of estimating the probability of an element manufactured on the assembly line being able to pass the quality tests.
- Integrating and automating the intelligent prediction system within the assembly line.
- Creation of an intuitive interface that allows parameterisation of the solution.
- Creation of a dashboard to help monitor the operation of the solution.
The DigiQual solution can be integrated into any type of company in which there is an assembly line and it is necessary to carry out quality tests on the products manufactured. DigiQual will use the historical data of each product and the results obtained in the quality tests to create prediction models capable of estimating the probability that other products with the same characteristics will pass these tests. The project will incorporate: Innovation in Artificial Intelligence, Innovation in Workflow Automation; and Innovation in Industrial Implementation of Artificial Intelligence Based Systems. Finally, it should be noted that the company is already in contact with several companies in the automotive sector that are interested in the results of the project and in testing the resulting prototype of the system.
With the financial support of:
