An improved test process model for cost reduction

Autor: James H. Mosher, Suresh Goyal
Rok vydání: 2006
Předmět:
Zdroj: Bell Labs Technical Journal. 11:173-190
ISSN: 1089-7089
DOI: 10.1002/bltj.20151
Popis: Testing is usually the most expensive, complicated, and time-consuming component of telecommunications equipment manufacture. In this paper, we present a mathematical model of the test process that allows the test-strategist to understand the non-intuitive relations between test-costs and benefits. The model is versatile and able to capture, in great detail, the process for the most complicated test and repair scenarios. Model inputs include test/repair/analysis times, yields, shipping and storage delays, and improvements in the above with learning. The expected time spent by a unit in the test process, and the associated cost, is derived in closed form. Unlike previous work, the solution is exact in modeling all the looping that can occur during test and repair, thereby improving accuracy and simplifying the mathematics. The model has been used for capacity planning, test asset allocation, NPI scheduling, cost reduction, yield improvement, alternate test plan comparisons, test optimization, and test strategy refinement.
Databáze: OpenAIRE