849804_en-US

liliconcom

Using Monte Carlo simulation to establish yield improvement decision indicators_Jeffrey Yang

Speaker: Jeffrey Yang, Senior Engineering Manager at Micron Semiconductor

Speech topic: Using Monte Carlo simulation to establish yield improvement decision indicators

summary:

In the LSI test industry, there are always many challenges to improve yield by reducing false failures, especially in back-end operations. On the one hand, customers’ requirements for on-time delivery are increasing. On the other hand, there is great pressure to save costs on each die while meeting customers’ quality standards. Finding a balance indicator has become a key point to guide engineering teams in making decisions. Monte Carlo simulation can effectively solve this problem, but engineers usually lack the appropriate tools to adopt this approach. JMP Pro can help you, even if you are not familiar with coding and algorithms. This article will introduce how to use JMP Pro to perform Monte Carlo simulation to find the right production tool recipe (socket disable rate, etc.) and operational KPIs (UPH, yield, recovery rate, etc.). The simulation results will be reflected in the radar chart to provide engineers with a dynamic view so that they can judge the balance point and adjust the recipe as needed. By formulating a series of radar indicators combined with the product portfolio and performing recipe adjustments, the overall yield of terminal production can be improved.