Description A Digital Twin for manufacturing relies heavily on data. Data are gathered from the design stage (modelling of a product, process and/or machine), the operation of machines and the final verification of a product. Thus, data need to be properly categorized and arranged in a database that can be easily maintained and queried. This database will be expanding over time and may need to accommodate more data, making flexibility, adaptability equally important to handling of big data for real-life industrial applications. AFRC is working on a Digital Twin on Forging of Titanium alloys and has constructed a SQL version of a database based on postgreSQL. The collected data are structured, but the variety of forging operations and sources of data pose a challenge for the efficient and effective management of data. The fixed format of SQL databases can thus become a limiting factor for a Digital Twin in practise. Therefore, it is of interest to examine the potential use of noSQL databases, like wide-column key-values databases (see Cassandra) and document-based database (see MongoDB JSON databases). Other alternatives like MySQL or postgreSQl with JSON entries could be examined if there is enough justification. Scalability of the database and ability to handle and query large datasets is of primary importance. Sample datasets will be provided by AFRC to assist the down-selection and design of the database as well as the schema of the existing postgreSQL database. The student will have the opportunity to work as part of a larger project team for the EPSRC DMWL project and will gain significant real-life experience on data management and industry 4.0 challenges in manufacturing. Key Objectives

1. Understanding of Digital Twim Infrastructure
2. Detailed data analysis and storage
3. Detail database management