This thesis reports on the application of Wireless Sensor Networks (WSNs) to the Single Point Diamond Turning (SPDT) process in order to monitor cutting tool wear and measure employee exposure to Volatile Organic Compounds (VOCs). WSNs offer a number of advantages over wired sensing technologies such as rapid deployment, low cost, and increased spatial resolution. These properties are useful for a number of sensing applications in manufacturing environments, and the information acquired by these systems can enable additional insights into manufacturing processes not achievable using other sensing methods.

SPDT is an extremely precise manufacturing process used for producing structures with accuracies of less than 1 µm and surface roughness' of less than 100 nm. The condition of the cutting tool is crucial to ensuring a good surface finish, however detecting tool wear in real time is challenging. A wireless sensor node to monitor the condition of the cutting tool in SPDT was successfully developed and implemented, based on low-cost, readily commercially available components. This system used an accelerometer to measure the vibration of the cutting tool, and was easily installed on existing SPDT machinery. A correlation between cutting tool wear and statistical features computed from the recorded vibration data was established and used to develop machine learning regression models to automatically determine the condition of the cutting tool in real-time. A number of regression models were compared, with neural network and Gaussian process regression algorithms performing best. These models were successfully implemented on the sensor node, enabling autonomous determination of the tool condition. Combined with wireless communication, trained regression models deployed on to the developed sensor node could drastically reduce wasted machining time and material in the SPDT process.

VOCs are organic compounds that take the form of a vapour at room temperature, and are emitted by solvents and cutting fluids used in the SPDT process. Monitoring exposure to VOCs is important to ensuring employee safety and confirming that protective measures are operating correctly. This work details the development of a WSN using Photo-Ionisation Detectors (PIDs) to quantify VOC concentrations in manufacturing environments. A WSN was developed and deployed in a research cleanroom to monitor emissions from spin coating, and in a SPDT environment to quantify machine operator exposure to VOC emissions from cutting fluids. The high spatial resolution offered by the WSN was found to be valuable in determining employee VOC exposure, as the VOC concentrations in the monitored environments exhibited highly localised behaviour. To further enable the VOC exposure of individual employees to be determined, the application of alternative sensing technologies to monitoring VOCs in the SPDT environment was explored. Chemiresistive sensors comprising of carbon nano tubes deposited onto interdigitated electrodes were developed. These sensors offered a significant improvement over some commercially available alternatives such as PIDs and metal oxide sensors, including a much smaller footprint (0.4 cm³) and greatly reduced power consumption (157.5 µW). The sensors exhibited good sensitivity and a high signal-to-noise ratio (104.84) in a laboratory setting. These sensors were novelly evaluated by comparison to a PID in an SPDT environment, and found to display good sensitivity in this setting, responding to ethanol concentrations of less than 1 ppm.