This research work investigates the development and application of process fingerprinting for conventional micro-injection moulding and ultrasonic micro injection moulding manufacturing of microneedle arrays for drug delivery. The process fingerprinting method covers in-depth analysis, interrogation and selection of certain process data features and correlation of these features with product fingerprints which are defined by the geometrical outcomes of the microneedle arrays in micro scale. The method was developed using the data collected using extensive sensor technologies attached to the conventional and ultrasonic micromoulding machines. Moreover, a machine vision based microneedle product evaluation apparatus is presented. Micromachining capabilities of different processes is also assessed and presented where state-of-the-art laser machining was used for microneedle tool manufacturing in the work. By using process fingerprinting procedures, conventional and ultrasonic micromoulding processes has been characterised thoroughly and aspects of the process that is affecting the part quality was also addressed for microneedle manufacturing. It was found that polymer structure is of paramount importance in obtaining sufficient microneedle replication. An amorphous polymer have been found to be more suitable for conventional moulding whereas semi-crystalline materials performed better in ultrasonic micromoulding. In-line captured micromoulding process data for conventional and ultrasonic moulding provided detailed insight of machine dynamics and understanding. Linear correlations between process fingerprints and micro replication efficiency of the microneedles have been presented for both micromoulding technologies. The in-line process monitoring and product quality evaluation procedures presented in this work for micro-injection moulding techniques will pave ways for zero-defect micromanufacturing of miniature products towards Industry 4.0.