Công cụ quản lý dự án Design Of Experiments DOE trong PMP là gì?
Công cụ này sử dụng trong quy trình: 8.1
A statistical method for identifying which factors may influence specific variables of a product or process under development or in production. DOE may be used during the Plan Quality Management process to determine the number and type of tests and their impact on cost of quality.
DOE also plays a role in optimizing products or processes. DOE is used to reduce the sensitivity of product performance to sources of variations caused by environmental or manufacturing differences. One important aspect of this technique is that it provides a statistical framework for systematically changing all of the important factors, rather than changing the factors one at a time. Analysis of the experimental data should provide the optimal conditions for the product or process, highlight the factors that influence the results, and reveal the presence of interactions and synergy among the factors. For example, automotive designers use this technique to determine which combination of suspension and tires will produce the most desirable ride characteristics at a reasonable cost.
#PS578 – A major financial institution has asked you to develop a cheque reconciliation application. The application will need to process over 1 million cheques a day. You need to process these in a nightly batch window of only 6 hours. You meet with your application architects and discuss various configuration scenarios – i.e. server size, CPU requirements, DASD and network dependencies. Your goal is to optimize performance in order to meet the 6 hour window time constraint. Which technique are you applying?
C. Analytical design
Design of experiments uses statistical methods to balance certain factors resulting in the most favourable outcome. It is a systematic approach to investigation of a system or process. A series of structured tests are designed in which planned changes are made to the input variables of a process or system. The effects of these changes on a pre-defined output are then assessed.