Controlling uncertainties is a challenging aspect in design and manufacturing of microsystems. As microsystems are characterised by features in the micro domain, product development and manufacturing processes are applied at the boundaries of their operational areas. In combination with many disciplines (mechanical, electrical, software, chemical etc.) and little standardisation, it causes microsystems development to be more time and cost intensive than products in the macro domain. Development of microsystems benefits from a concurrent approach of product and production design. Uncertainties may be addressed by application of methods for systems engineering (engineering design). Systems engineering applies models for the analysis of projects, usually a linear set of gates that need to be closed successively as the project evolves. Over the last ten years, models with an iterative approach of design and testing, gained in popularity due to their more agile characteristic that performs better in fast changing markets. Microsystems development benefits from the linear approach that performs well for their structured project control, but because of the high market dynamics, agile methods will speed up the process, which results in faster market introduction, advances the product life cycle, and increases return on investments. Currently, there are no known systems engineering models that combine linear and iterative monitoring of projects to gain the best of both methods, especially not in combination with the capability of concurrently monitoring the development of product and production design. This thesis investigates how existing ways of system engineering can be combined to: (RQ1) enable iterative and linear modelling of microsystems development, and (RQ2) merge these qualities into a combined model to monitor the development process concurrently. The first problem is addressed by (RQ1): i. Modelling development progression by execution of iterative cycles that alternately perform functional system decomposition and functional gating. ii. This iterative model is elevated with the method of Axiomatic Design to enable concurrent system decomposition. Implementation of elements from the V-Modell XT enable functional gating to index the concurrent development process iii. The ‘Theory of Complexity’ of Axiomatic Design is applied to realise an intelligent, knowledge based, gating function to be used as a continuous maturity measure; The results show that linear and iterative models can be merged successfully. With some extensions, the Theory of Complexity of Axiomatic Design can indeed be used for continuous monitoring of product and process development. The thus-obtained maturity measure can be applied for the analysis of project decisions. This was successfully done for retrospective analysis of two cases. To merge the qualities of analyses ‘i to iii’ into a combined model to monitor the development process concurrently, three tools for application have been developed (RQ2). iv. The first is a method for visualisation of the intelligent gating function, based on analysis ‘iii’. The method applies a newly developed ‘Maturity Diagram’ that plots the Design Axioms as continuous parameters v. The second is a method for assessment of reconfigurable manufacturing systems based on analysis ‘ii’. The method estimates the investigations needed to (re)configure a product specific manufacturing system vi. The third is a tool for roadmapping and monitoring that combines outcomes of analyses ‘i, ii, and iii’. This model is called ‘Constituent Roadmap’ and it is based on: (a) an iterative approach, (b) concurrent decomposition, (c) the advanced gating function, and (d) knowledge application to the product and process design. The Constituent Roadmap was applied for the development of a ‘smart dust’ sensor system. It was found to structure knowledge development and application. This increases the chances to satisfy the functional requirements of the design. In parallel, it functions as a communications tool between designers and managers. Together, a reasonably complete picture has emerged how the design of microsystems and their production means can be modelled, and how uncertainties may be categorised so they can be addressed in the best order.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:759635 |
| Date | January 2017 |
| Creators | Puik, Erik |
| Publisher | University of Warwick |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://wrap.warwick.ac.uk/104237/ |
Page generated in 0.0035 seconds