Evaluating the Impacts of Reticle Requirements in Semiconductor Wafer Fabrication
S.L.M. Diaz, John Fowler, Michele E. Pfund, Gerald T. Mackulak, Matt Hickie
Abstract
S.L.M. Diaz, John Fowler, Michele E. Pfund, Gerald T. Mackulak, Matt Hickie
Abstract
Photolithography is generally the most constraining workstation in a semiconductor fabrication facility. Up to this point, much of the research and analysis in this area has not included the inherent system requirements created by reticle masks, but the fact remains that in order to process a job in the photolithography station, the job must be ready, the machine must be idle, and the reticle must be inspected and setup on the idle machine. To better understand the impacts of the reticle requirements in the production environment, a discrete-event simulation model of the photolithography station has been created and coupled with a network flow optimization model that optimizes the location of all reticles at 6-h intervals. All other production processes are modeled using a generic stochastic processing delay. Using this framework and a 2/sup 6/ full factorial-designed experiment, we have identified when optimal reticle management processes positively impact the productivity of the overall facility the most. These results were then used to derive and test new dispatching policies for the photolithography workstation.
OpenAlex reports 34 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Photolithography is generally the most constraining workstation in a semiconductor fabrication facility. Up to this point, much of the research and analysis in this area has not included the inherent system requirements created by reticle masks, but the fact remains that in order to process a job in the photolithography station, the job must be ready, the machine must be idle, and the reticle must be inspected and setup on the idle machine. To better understand the impacts of the reticle requirements in the production environment, a discrete-event simulation model of the photolithography station has been created and coupled with a network flow optimization model that optimizes the location of all reticles at 6-h intervals. All other production processes are modeled using a generic stochastic processing delay. Using this framework and a 2/sup 6/ full factorial-designed experiment, we have identified when optimal reticle management processes positively impact the productivity of the overall facility the most. These results were then used to derive and test new dispatching policies for the photolithography workstation.
Key concepts: Reticle, Photolithography, Semiconductor device fabrication, Wafer fabrication, Workstation, Wafer, Discrete event simulation, Reliability engineering