Manufacturing system testing measurement and management process.

摘要

This dissertation will address the key question of how a test engineer measures, manages, and improves the quality of a system level manufacturing test process. The question is complicated due to the lack of clear metrics to drive improvements to the test process. It is an interesting question because of the way it blends test engineering and process management techniques. This dissertation will examine the question using a combination of several process management tools and test engineering techniques. The management processes include Total Quality Management, the Balanced Scorecard, and the Theory of Constraints.; The methodology used to address the issues identified, follows the standard Total Quality Management six-sigma process of Define, Measure, Analyze, Improve, and Control. The define phase will answer the question "How do you know if you have a good test process?" This is achieved by defining the stakeholder requirements for the test process. A balanced scorecard for the test process will be the result of the define phase. The measurement phase will outline how these requirements are measured. This includes defining test coverage at the system level, and creating models that calculate the quality, capacity, and cost impact of the test process. The cost models showed that a significant cost is related to test development and support. The analysis phase combines the Balanced Scorecard and the Theory of Constraints to identify the core conflict leading to the low efficiency of the test process. The core conflict identified is a cost versus quality tradeoff.; The improve phase seeks to eliminate this core conflict by proposing that the test process be designed as tool for quality improvement, not just a quality screen. This approach has several implications on how systems are tested and requires a more communicative process. In combination with the need to reduce the cost of test development, these implications lead the author to propose a new Built in Self Test standard for system testing. The control phase will therefore, discuss the projected results including a test time saving of 50% and a cost saving of a similar proportion.