Dimensional measurement is fundamental to ensuring product quality and improving manufacturing process efficiency. Traditional tactile probing methods are recognised for their accuracy but are limited by slow data acquisition. Conversely, non-contact systems offer higher speed and higher data density but introduce important considerations regarding measurement uncertainty and data reliability. This research seeks to establish a robust framework for the early adoption of advanced metrological systems characterised by unitless control parameters, with particular emphasis on integrating laser line scanners into coordinate measuring machines (CMMs). The study systematically quantifies measurement uncertainty, optimises scanner settings, and compares non-contact and tactile measurements to evaluate the reliability and industrial applicability of non-contact metrology technologies. Investigations were undertaken using an LK Evolution CMM equipped with both tactile and laser probing systems. Both Full Factorial and Taguchi Design of Experiments (DoE) methodologies were employed to understand the interactions among unitless scanner parameters. Data processing was carried out utilising MATLAB algorithms and PolyWorks software for point cloud alignment and deviation assessment. An uncertainty budget was established in accordance with ISO 15530-3. Optimal scanner settings were identified, reducing variability and improving point cloud quality. Taguchi designs produced results comparable to Full Factorial designs with fewer runs. The uncertainty budget for the LC15Dx scanner was ±0.0375 mm (k = 2). No systematic errors were detected after filtering and alignment, supporting the feasibility of integrating non-contact measurements. The study demonstrates that a structured framework incorporating DoE and uncertainty analysis enables reliable optimisation of non-contact metrology systems. Laser line scanners, when properly configured, can deliver measurement performance within acceptable uncertainty limits. The proposed framework accelerates organisational learning, reduces risk in early adoption, and provides a transferable methodology for other advanced metrology technologies. This approach supports industry goals of improving throughput without compromising quality, offering practical guidance for integrating unitless parameter-based systems into manufacturing environments.
| Date of Award | 4 Mar 2026 |
|---|
| Original language | English |
|---|
| Sponsors | National Physical Laboratory |
|---|
| Supervisor | Andrew Longstaff (Main Supervisor) & Simon Fletcher (Co-Supervisor) |
|---|