Public Defence in Mechanical Engineering M.Sc. (Tech) Tekin Uyan

"Quality assurance through advanced manufacturing in metal casting foundries"

SMART FOUNDRY: The foundry industry has significant environmental concerns such as air and solid waste pollution, and depletion of natural resources. On the other hand, metal casting is used in a wide range of essential products, from wind turbine torque arms to electric car's wheels. Since earth's metal resources are limited, so recycling is necessary but one bottleneck is that the recycling industry has difficulty tracing the metal waste. Moreover, metal casting is quite costly to the foundries. Operating heavy-duty industrial machinery, like smelters that melt the ores at scorching temperatures above 1000 degrees Celsius, takes a massive amount of energy and human effort. Also, the quality requirements are increasing by huge scale of production and growing industrial demand for more specific metal products. Defect reduction and root cause elimination in metal castings is necessary to save energy, reduce rework and scrap, and meet industry standards. It also helps reduce heavy industry's negative impact on the environment and foundry workers (aka: Foundrymen). This research addresses three main problems that have been identified as obstacles to the industrial 4.0 transformation of the metal foundry industry: providing a means of creating permanent, digitally-readable markings on cast parts, improving the way data is collected and associated with those parts, and developing new methods for statistical quality control using the data collected. The results of the study show that an operator can uniquely mark and track the parts throughout the foundry with a new developed method using additively manufactured (AM) marking tags. This application enables a smart foundry quality management system: simple digital tracking operation via mobile phones utilizing readily available barcode reader applications. Additionally, using the data collected from the system, the researcher was able to create machine learning (ML) model that can predict when cast parts will be defective, which can help in pre-series production for new products. This research brings new information to the field by showing that it is possible to use technology to track cast parts and collect data on the production process a way that makes it easier to identify and address quality control issues. The information from this research can be applied in metal foundries by using AM tags to digitally mark and track cast parts, collect data on the production process, and create ML models to predict defects.