Boğaziçi University Data Science and Artificial Intelligence Institute faculty members Assoc. Prof. Dr. Şener Özönder and Assoc. Prof. Dr. Hüseyin Oktay Altun, a faculty member at Boğaziçi University Institute of Data Science and Artificial Intelligence, and Assoc. Prof. Dr. Berk Ayvaz, a faculty member at Istanbul Commerce University, as executives and Assoc. Prof. Dr. Berk Ayvaz as a consultant, the “Artificial Intelligence Supported Cost Estimation Software Development Project from Technical Drawings” aims to accelerate the cost estimation and appropriate pricing policy determination processes for the products that enterprises plan to produce with machine learning.
Financial loss and time loss will be reduced
Sharing the information that the project, which will be supported by TÜBİTAK 1711 Artificial Intelligence Ecosystem Call, which is among the competitive programmes of TÜBİTAK, will last 18 months. Prof. Dr. Hüseyin Oktay Altun expressed the process of the emergence of the project idea as follows: “It can take a long time for companies to determine the prices of the products they produce and will produce. However, price offers resulting from faulty cost analyses also affect companies financially. The software we have developed leaves this job to machine learning. Thus, it becomes possible to minimise financial losses and time loss.”
Stating that all relevant data of the company will be analysed with artificial intelligence algorithms for the price determination of the software developed within the scope of the project, Dr. Prof Altun continued as follows: “The software will automatically perform the cost estimation that the company wants to make for its new or already marketed product by examining the technical drawings of the products in the company. Thanks to machine learning, production processes will be improved. This solution will guide companies and accelerate decision-making processes.”
Artificial intelligence ecosystem gains momentum in Turkey
Prof. Dr. Şener Özönder, Associate Professor at Boğaziçi University Data Science and Artificial Intelligence Institute, stated that the TÜBİTAK 1711 Artificial Intelligence Ecosystem Call has assumed a very critical role in Turkey in terms of both the product development of initiatives operating in the field of deep technology and meeting the need for effective solutions based on artificial intelligence in the public and private sectors. He also said the following about the cooperation established for the machine learning supported software to be developed: “This project will be a joint university-industry cooperation project of the Institute of Data Science and Artificial Intelligence, where I am also a faculty member, my initiative ArtificaX Informatics and Teknorot, which produces vehicle steering and suspension parts, which will be the customer of the software we develop. Our goal is to develop a multimodal machine learning algorithm that will simultaneously process geometry, text and part images from the technical drawings of the parts used in the production phase and cost estimation. The software embedded with this algorithm will take the technical drawing of the product to be produced from the ERP system of the enterprise. With the artificial intelligence algorithm, it will estimate quantities such as cost estimation, energy use, production time and carbon emissions and transmit the results back to the ERP system. Thus, cost estimation analyses, which normally take up to 10 days when done manually, will be reduced to 60 seconds. The aim here is both to reduce the cost of the business in these processes and to increase the global competitiveness of the business with fast cost estimation and customer proposal creation. The software we have developed will produce benefits for every industry where technical drawing is used in production, from automotive to aviation.”
