Additive Manufacturing (AM) offers freedom in both design and materials selection. Safety critical industries, such as aerospace and healthcare, stand to benefit from the unique capabilities offered by AM. However, adoption of AM is hindered by the unique and complex inspection and quality assurance requirements that conventional line-of-sight techniques struggle to fulfil. X-ray computed tomography (XCT) is a non-destructive, non-line-of-sight, volumetric imaging technique, which has gained traction as a viable inspection method over the last two decades and shows promise as a next generation dimensional metrology tool for AM. This thesis details the development of a system of modular artefacts which provide a reconfigurable toolkit to address a variety of AM metrology challenges. The toolkit consists of AM test, and XCT data validation modules. The AM test modules are engineered to assess the ability of an AM system to produce challenging and industrially relevant geometries. The modular system also includes a number of XCT data validation modules, which are intended to reduce measurement uncertainty by providing a consistent, repeatably measured ground truth in the form of features of known size and shape in each dataset. The AM test modules are parameterised based on AM process and XCT scanning parameters, enabling the modules to be adapted for specific requirements. The modular system is developed using a combination of Design for Metrology (DfM) and Design for Additive Manufacturing (DfAM) methods. Both methods consider the benefits and limitations of the respective systems during the design process, resulting in features that are optimised for the specific manufacturing and metrology systems. This thesis establishes a foundation for future development of the modular artefact system. The modular artefact family could be extended to additional AM modalities and materials, including wire‑based directed energy deposition (DED) which is often used for near‑net‑shape manufacture. Further work could also enhance the XCT data‑quality modules to represent as‑built AM surface texture characteristics, incorporating the influence into uncertainty estimation. Through such developments, the modular system could evolve into a broadly applicable benchmark for both research and industrial workflows, supporting robust, repeatable, and traceable AM–XCT quality assurance.
| Date of Award | 7 Sept 2026 |
|---|
| Original language | English |
|---|
| Sponsors | National Physical Laboratory & Engineering and Physical Sciences Research Council |
|---|
| Supervisor | Shan Lou (Main Supervisor) & Wenhan Zeng (Co-Supervisor) |
|---|