AI-CFPD - Conceptual Fashion Product Design Assisted by Artificial Intelligence

The project aims at facilitating the design of clothing products, particularly in the field of product concept development, by providing personalized proposals to the designer as inspiration, by developing tool based on Artificial Intelligence methods. The tools will work in addition to existing support systems for the development of clothing products such as Life Cycle Management Systems, CAD (Graphics), 3D Modeling systems, commonly used by companies and designers.

In the clothing sector, design, development and procurement teams have been affected more than any other industry and are constantly being under pressure to present more products with fewer resources in a shorter time. The diversity of garment designs created as new products is not found in any other industry and is almost independent of the size of the business. The factors that influence the creation of new products and fashion prediction are constantly increasing and processing is taking place in a shorter time. Businesses have been forced to look at their internal processes and reassess the way they bring their products to the market and integrate new technologies.

The proposed research will move in two dimensions: The first dimension concerns the perfect transformation of the product design field into a field of action planning that can be supported by Artificial Intelligence methods, providing timely and valid information to the designer drawing information from a wider range of sources than today's method. The priority areas for the fashion sector will be performed in cooperation with the participating companies aiming to serve them as well as to create a product suitable for all companies in the sector.

The second dimension of the research concerns the design and implementation of an intelligent and semi-autonomous decision support system for fashion designers. The system proposed in this proposal consists of 4 individual subsystems, the Personnel Assistant Subsystem, the Business Rules Subsystem, the Data Analysis Subsystem and the Visualization Subsystem. The system will include a Reinforced Learning process in which the end user will be able to give feedback on system choices: he will reward good choices while punishing options he does not like. For the interconnection of the individual subsystems, a modular architecture will be followed, which will give flexibility in case of changing / removing some sub-segments or adding future extensions.

The consortium consists of 2 academic institutions and 3 companies, all of them SMEs. Computer Life is the Coordinator, as an IT company offering solutions to the Textile/Clothing sector, responsible for the exploitation and commercialization of the product. The two universities are complementary to the scope of the proposal. The CAD laboratory of the Technical University of Crete holds experience in the design of clothing products and the ISSEL laboratory of the Aristotle University of Thessaloniki in the Field of Artificial Intelligence. The two end-user companies, Abotis and Energiers, have strong clothing development departments and will be specifications providers for product development and evaluators of the developed product.

The project consists of 7 Work Packages (WPs), where all partners are involved with distinctive roles. The results of the project will have an impact on all companies in the apparel industry, disseminated through publications, announcements and visits to other companies. The commercial exploitation of the software will be promoted by Computer Life.

This research has been co-financed by the European Union and Greek national funds through the Operational Program Competitiveness, Entrepreneurship and Innovation, under the call RESEARCH-CREATE-INNOVATE (project code: T1EDK-03464)

Total funding: 970.925,00 €