• 제어로봇시스템학회논문지
• /
• 제6권1호
• /
• pp.104-111
• /
• 2000

A fuzzy logic Run-by-Run(RbR) controller and an in -line wafer characteristics prediction scheme for the rapid thermal processing system have been developed for the study of process repeatability. The fuzzy logic RbR controller provides a framework for controlling a process which is subject to disturbances such as shifts and drifts as a normal part of its operation. The fuzzy logic RbR controller combines the advantages of both fuzzy logic and feedback control. It has two components : fuzzy logic diagnostic system and model modification system. At first, a neural network model is constructed with the I/O data collected during the designed experiments. The wafer state after each run is assessed by the fuzzy logic diagnostic system with featuring step. The model modification system updates the existing neural network process model in case of process shift or drift, and then select a new recipe based on the updated model using genetic algorithm. After this procedure, wafer characteristics are predicted from the in-line wafer characteristics prediction model with principal component analysis. The fuzzy logic RbR controller has been applied to the control of Titanium SALICIDE process. After completing all of the above, it follows that: 1) the fuzzy logic RbR controller can compensate the process draft, and 2) the in-line wafer characteristics prediction scheme can reduce the measurement cost and time.

• Industrial Engineering and Management Systems
• /
• 제9권3호
• /
• pp.262-274
• /
• 2010

This article investigates how to adaptively predict the time-varying metrology delay that could realistically occur in the semiconductor manufacturing practice. Metrology delays pose a great challenge for the existing run-to-run (R2R) controllers, driving the process output significantly away from target if not adequately predicted. First, the expected asymptotic double exponentially weighted moving average (DEWMA) control output, by using the EWMA and recursive least squares (RLS) prediction methods, is derived. It has been found that the relationships between the expected control output and target in both estimation methods are parallel, and six cases are addressed. Within the context of time-varying metrology delay, this paper presents a modified recursive least squares-linear trend (RLS-LT) controller, in combination with runs test. Simulated single input-single output (SISO) R2R processes subject to various time-varying metrology delay scenarios are used as a testbed to evaluate the proposed algorithms. The simulation results indicate that the modified RLS-LT controller can yield the process output more accurately on target with smaller mean squared error (MSE) than the original RLSLT controller that only deals with constant metrology delays.

• Industrial Engineering and Management Systems
• /
• 제4권1호
• /
• pp.54-67
• /
• 2005

This paper provides an evaluation of an optimization-based, multiple-input double-output (MIDO) run-to-run (R2R) control scheme for general semiconductor manufacturing processes. The controller in this research, termed adaptive dual response optimizing controller (ADROC), can serve as a process optimizer as well as a recipe regulator between consecutive runs of wafer fabrication. In evaluation, it is assumed that the equipment model could be appropriately described by a pair of second-order polynomial functions in terms of a set of controllable variables. Of practical relevance is to consider a drifting effect in the equipment model since in common semiconductor practice the process tends to drift due to machine aging and tool wearing. We select a typical application of R2R control to chemical mechanical planarization (CMP) in semiconductor manufacturing in this evaluation, and there are five different CMP process scenarios demonstrated, including mean shift, variance increase, and IMA disturbances. For the controller, ADROC, an on-line estimation technique is implemented in a self-tuning (ST) control manner for the adaptation purpose. Subsequently, an ad hoc global optimization algorithm based on the dual response approach, arising from the response surface methodology (RSM) literature, is used to seek the optimum recipe within the acceptability region for the execution of next run. The main components of ADROC are described and its control performance is assessed. It reveals from the evaluation that ADROC can provide excellent control actions for the MIDO R2R situations even though the process exhibits complicated, nonlinear interaction effects between control variables, and the drifting disturbances.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2005년도 ICCAS
• /
• pp.519-524
• /
• 2005

A numerical process to simulate SiO2 dry etching with inductively coupled C2F6 plasmas has been constructed using a commercial CFD code as a first step to design a run-to-run control system. The simulator was tuned to reasonably predict the reactive ion etching behavior and used to investigate the effects of plasma operating variables on the etch rate and uniformity. The relationship between the operating variables and the etching characteristics was mathematically modeled through linear regression for future run-to-run control system design.

• International Journal of Precision Engineering and Manufacturing
• /
• 제9권1호
• /
• pp.12-18
• /
• 2008

This paper describes two run-to-run controllers, a nonlinear multiple exponential-weight moving-average (NMEWMA) controller and a dynamic model-tuning minimum-variance (DMTMV) controller, for photolithography processes. The relationships between the input recipes (exposure dose and focus) and output variables (critical dimensions) were formed using an experimental design method, and the photolithography process model was built using a multiple regression analysis. Both the NMEWMA and DMTMV controllers could update the process model and obtain the optimal recipes for the next run. Quantified improvements were obtained from simulations and real photolithography processes.

• 응용통계연구
• /
• 제31권4호
• /
• pp.485-495
• /
• 2018

CCC-r 관리도는 고품질공정에서 공정불량률을 관리하는 경우 매우 효율적이라고 알려져 있다. 이 관리도를 사용할 때 관리상태의 공정모수는 일반적으로 알려져 있지 않기 때문에 제1국면의 표본을 추출하여 이를 추정해야 한다. 제2국면에서 관리도의 성능은 제1국면에서 추정한 모수와 관리한계에 영향을 받기 때문에, 추정 오차의 영향을 살펴보는 것은 중요하다. 이 논문에서는 일반적으로 많이 사용하는 평균런길이의 평균(average of average run length) 이외에 평균런길이의 표준편차(standard deviation of average run length)를 사용하여 CCC-r 관리도의 관리상태의 성능을 평가하였다. 그 결과 CCC-r 관리도에서 안정적인 관리상태의 성능을 유지하기 위해서는 이전에 권장하던 제1국면의 표본 크기보다 훨씬 더 큰 표본이 필요하다는 사실을 알 수 있었다.

• Industrial Engineering and Management Systems
• /
• 제4권2호
• /
• pp.158-164
• /
• 2005

Kusukawa and Ohta presented the $CS_{CQ-r}$ chart to monitor the process defect $rate{\lambda}$ in high-yield processes that is derived from the count of defects. The $CS_{CQ-r}$ chart is more sensitive to $monitor{\lambda}$ than the CQ (Cumulative Quantity) chart proposed by Chan et al.. As a more superior chart in high-yield processes, we propose a Synthetic chart that is the integration of the CQ_-r chart and the $CS_{CQ-r}$chart. The quality characteristic of both charts is the number of units y required to observe r $({\geq}2)$ defects. It is assumed that this quantity is an Erlang random variable from the property that the quality characteristic of the CQ chart follows the exponential distribution. In use of the proposed Synthetic chart, the process is initially judged as either in-control or out-of-control by using the $CS_{CQ-r}$chart. If the process was not judged as in-control by the $CS_{CQ-r}$chart, the process is successively judged by using the $CQ_{-r}$chart to confirm the judgment of the $CS_{CQ-r}$chart. Through comparisons of ARL (Average Run Length), the proposed Synthetic chart is more superior to monitor the process defect rate in high-yield processes to the stand-alone $CS_{CQ-r}$ chart.

• JSTS:Journal of Semiconductor Technology and Science
• /
• 제11권1호
• /
• pp.1-5
• /
• 2011

To enter next process control, numerous approaches, including run-to-run (R2R) process control and fault detection and classification (FDC) have been suggested in semiconductor manufacturing industry as a facilitation of advanced process control. This paper introduces a novel type of optical plasma process monitoring system, called plasma eyes chromatic system (PECSTM) and presents its potential for the purpose of fault detection. Qualitatively comparison of optically acquired signal levels vs. process parameter modifications are successfully demonstrated, and we expect that PECSTM signal can be a useful indication of onset of process change in real-time for advanced process control (APC).

• 생물환경조절학회지
• /
• 제24권3호
• /
• pp.153-160
• /
• 2015

본 연구는 환경오염과 양수분 손실을 주는 비순환식 수경재배에 FDR센서를 이용한 자동관수시스템을 적용할 때 관수효율을 높이기 위한 최적의 최소대기시간을 설정하고자 수행되었다. 실험은 가을과 겨울철에 봄과 여름철에 두 번 수행하였고 가을과 겨울철에는 3분 급액과 최소대기시간을 5분으로 한 3R5F 처리구, 3분 급액과 최소대기시간을 10분으로 한 3R10F 처리구, 5분 급액과 최소대기시간을 15분으로 한 5R15F 처리구를 설정하여 실험하였고 봄과 여름철에는 3분 급액과 최소대기시간을 5분으로 한 3R5F 처리구, 3분 급액과 최소 대기시간을 10분으로 한 3R10F 처리구를 설정하여 실험하였다. 3분 급액은 주당 60mL, 5분 급액은 주당 80mL가 공급되었다. 가을과 겨울철 재배에서 정식 후 62일 까지 주당 급액량은 3R5F (858mL) > 5R15F (409mL) > 3R10F (306mL) 처리 순으로 나타났고 배액률은 3R5F (44%) > 5R15F (23%) > 3R10F (14%) 순으로 나타났다. 정식 후 62일부터 102일 까지는 일일 주당 급액량이 5R15F (888mL)> 3R5F (695mL)> 3R10F (524mL) 순으로 나타났고 이 시기에 배액률은 5R15F에서 가장 높았다. 봄과 여름재배에서는 일일 주당 급액량과 배액율이 3R5F 처리구에서 3R10F 처리구보다 높았다. 두 재배 모두에서 수분이용효율 (WUE)은 3R10F 처리에서 높았다. 따라서 FDR 센서를 활용한 자동화 관수 시스템에서 관수효율을 높이기 위한 최소대기시간은 10분으로 고찰된다.

• 한국시뮬레이션학회논문지
• /
• 제16권2호
• /
• pp.37-44
• /
• 2007

EWMA(지수가중평균)을 사용한 제어 방법은 반도체 제조 공정의 런투런 제어 환경에서 사용되는 대표적인 제어 방법이다. EWMA 제어에서는 제어 모수 값의 선택이 제어 결과에 주된 영향을 미치기 때문에, 공정 상황에 적합한 제어 모수 값을 사용하는 것이 중요하다. 불안정적인 공정 상황에서는 변화하는 공정 상황에 적합한 값으로 EWMA 제어 모수를 동적으로 변경하면서 제어해 주는 것이 고정된 모수 값을 사용하는 EWMA 제어에 비해 효율적이다. 본 연구에서는 동적인 EWMA 제어 모수를 사용한 기존 연구들을 살펴보고 이들의 단점을 보완한 새로운 알고리즘을 제시하며, 시뮬레이션을 통해 우수한 수행 결과를 확인하였다.