A smart sensing platform for the discrimination of human movement

This thesis documents the design, implementation and testing of a smart sensing platform that is able to discriminate between differences or small changes in a persons walking. The distributive tactile sensing method is used to monitor the deflection of the platform surface using just a small number of sensors and, through the use of neural networks, infer the characteristics of the object in contact with the surface. The thesis first describes the development of a mathematical model which uses a novel method to track the position of a moving load as it passes over the smart sensing surface. Experimental methods are then described for using the platform to track the position of swinging pendulum in three dimensions. It is demonstrated that the method can be extended to that of real-time measurement of balance and sway of a person during quiet standing. Current classification methods are then investigated for use in the classification of different gait patterns, in particular to identify individuals by their unique gait pattern. Based on these observations, a novel algorithm is developed that is able to discriminate between abnormal and affected gait. This algorithm, using the distributive tactile sensing method, was found to have greater accuracy than other methods investigated and was designed to be able to cope with any type of gait variation. The system developed in this thesis has applications in the area of medical diagnostics, either as an initial screening tool for detecting walking disorders or to be able to automatically detect changes in gait over time. The system could also be used as a discrete biometric identification method, for example identifying office workers as they pass over the surface.

Searching for lean white-collar workers

The purpose of this research has been to investigate the extent to which Lean techniques have been applied to indirect or White Collar activities within manufacturing operations. The work has been based on a systematic review of the literature and six UK based case studies. Both are described in the paper. The findings from this work include, for example, that there are few reports off the application of Lean in this context. The more substantive papers focus on NPI and Project Management, where as, Support and ‘Front Office’ activities are covered only briefly in ‘trade journal’ articles. These do demonstrate that Lean techniques can be applied successfully to non-production related activities, and there are significant opportunities for work in this area. However, the definition of Lean is evolving and practitioners do not share a common understanding of terminology. This may undermine their accuracy when reporting the application of Lean techniques.