New Optimization Strategy for Chemical Mechanical Polishing Process.

Wang, Gou‐Jen; Chen, Jau-Liang; Hwang, Ju-Yi

doi:10.1299/jsmec.44.534

Cited by 6 publications

(6 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…From the 54 source data sets [8], [10] (as tabulated in Table XII of Appendix III), 45 data sets were randomly chosen, similar to the Case I in part A, to be the training data for training NN and ANFIS models and the other 9 data sets for testing the trained models. Two NN models (i.e., purelin and tansig) and two ANFIS models (i.e., GP and SC) were used to establish models of MRR and WIWNU, respectively.…”

Section: B Models For Case IImentioning

confidence: 99%

Adaptive Neuro-Fuzzy Inference System Modeling of MRR and WIWNU in CMP Process With Sparse Experimental Data

Lih¹,

Bukkapatnam

Rao

et al. 2008

IEEE Trans. Automat. Sci. Eng.

View full text Add to dashboard Cite

Availability of only limited or sparse experimental data impedes the ability of current models of chemical mechanical planarization (CMP) to accurately capture and predict the underlying complex chemomechanical interactions. Modeling approaches that can effectively interpret such data are therefore necessary. In this paper, a new approach to predict the material removal rate (MRR) and within wafer nonuniformity (WIWNU) in CMP of silicon wafers using sparse-data sets is presented. The approach involves utilization of an adaptive neuro-fuzzy inference system (ANFIS) based on subtractive clustering (SC) of the input parameter space. Linear statistical models were used to assess the relative significance of process input parameters and their interactions. Substantial improvements in predicting CMP behaviors under sparse-data conditions can be achieved from fine-tuning membership functions of statistically less significant input parameters. The approach was also found to perform better than alternative neural network (NN) and neuro-fuzzy modeling methods for capturing the complex relationships that connect the machine and material parameters in CMP with MRR and WIWNU, as well as for predicting MRR and WIWNU in CMP.Note to Practitioners-In many microelectronics and other industrial applications, the cost of experimentation tends to be prohibitively high. Consequently, only a limited or sparse set of data is available for modeling and optimization of the process. Under such circumstances, the present approach based on ANFIS with SC was found to perform better than the alternative NN or neuro-fuzzy modeling methods for capturing the complex relationships that connect the machine and material parameters in the process performance variables, such as the MRR and WIWNU in the CMP process. These are considered as critical factors for improving wafer yield.Index Terms-Adaptive neuro-fuzzy inference system (ANFIS), chemical mechanical planarization (CMP), neural network (NN).

show abstract

Section: B Models For Case IImentioning

confidence: 99%

Adaptive Neuro-Fuzzy Inference System Modeling of MRR and WIWNU in CMP Process With Sparse Experimental Data

Lih¹,

Bukkapatnam

Rao

et al. 2008

IEEE Trans. Automat. Sci. Eng.

View full text Add to dashboard Cite

show abstract

“…According to previous results [17], we chose five important and adjustable process parameters as control variables, namely down force, platen speed, solid content, back pressure, and polishing time.…”

Section: Processes Parametersmentioning

confidence: 99%

“…Due to their features of massive parallel computing and their ability to learn and generalize, neural networks have been applied to many engineering and science problems, including pattern classification, function approximation, nonlinear system modeling and control, image processing, and parameters optimization [14][15][16][17].…”

Section: Introductionmentioning

confidence: 99%

Developing a neural network-based run-to-run process controller for chemical-mechanical planarization

Wang

2006

Int J Adv Manuf Technol

Self Cite

View full text Add to dashboard Cite

A new neural network-based run-to-run process control system (NNRtRC) is proposed in this article. The key characteristic of this NNRtRC is that the linear and stationary process estimator and controller in the exponentially weighted moving average (EWMA) run-to-run control scheme are replaced by two multilayer feed-forward neural networks. An efficient learning algorithm inspired by the sliding mode control law is suggested for the neural network-based run-to-run controller. Computer simulations illustrate that the proposed NNRtRC performs better than the EWMA approach in terms of draft suppression and adaptation to environmental change. Experimental results show that the NNRtRC can precisely trace the desired target of material removal rate (MRR) and keep the within wafer non-uniformity (WIWNU) in an acceptable range

show abstract

“…According to the results from references [8] and [9], we chose four important and adjustable process parameters as control variables: down force, platen speed, solid content and back pressure. The relationship between removal rate and polishing pressure and polishing speed can be described by Preston equation [12]:…”

Section: Processes Parametersmentioning

confidence: 99%

“…In this study, based on the existing CMP process parameter optimisation techniques [8] and the mechanical wear mechanism of CMP [9], with the consideration of polishing time effect on removal rate and planarisation, we hope to develop a new CMP endpoint detection technique that doesn't require modification of current machine structure or any additional instruments. In other words, this new technique is able to end the polishing processes through polishing time control.…”

Section: Introductionmentioning

confidence: 99%

A neural-Taguchi-based quasi time-optimization control strategy for chemical-mechanical polishing processes

Wang

Chou

2005

Int J Adv Manuf Technol

View full text Add to dashboard Cite

Besides the major factors such as the down force, back pressure and the rotating speed of wafer carrier, effect of polishing time is also an important issue in CMP processes. In this study, a neural-Taguchi method based on a cost-effective quasi time-optimisation technique for chemical-mechanical polishing (CMP) processes is developed. The key concept of this new technique is that an optimal process parameter set is obtained through a neural-network-simulated CMP process model. Under such an optimal parameter set, the desired material removal rate within-wafer-nonuniformity can be reached with the optimal polishing time. It has been proved by experiment that the proposed method can offer a better polishing performance while reducing the polishing time by 1/3

show abstract

New Optimization Strategy for Chemical Mechanical Polishing Process.

Cited by 6 publications

References 6 publications

Adaptive Neuro-Fuzzy Inference System Modeling of MRR and WIWNU in CMP Process With Sparse Experimental Data

Adaptive Neuro-Fuzzy Inference System Modeling of MRR and WIWNU in CMP Process With Sparse Experimental Data

Developing a neural network-based run-to-run process controller for chemical-mechanical planarization

A neural-Taguchi-based quasi time-optimization control strategy for chemical-mechanical polishing processes

Contact Info

Product

Resources

About