Adaptive Metrology Sampling techniques enabling higher precision in variability detection and control

Test of covariance changes without a large sample and its application to fault detection and classification

Semiconductor manufacturing comprises hundreds of consecutive unit processes. A single misprocess could jeopardize the whole manufacturing process. In current manufacturing environments, data monitoring of equipment condition, wafer metrology, and inspection, etc., are used to probe any anomaly during the manufacturing process that could affect the final chip performance and quality. The purpose of investigation is fault detection and classification (FDC). Various methods, such as statistical or data mining methods with machine learning algorithms, have been employed for FDC. In this paper, we propose an artificial immune system (AIS), which is a biologically inspired computing algorithm, for FDC regarding semiconductor equipment. Process shifts caused by parts and modules aging over time are main processes of failure cause. We employ state variable identification (SVID) data, which contain current equipment operating condition, and optical emission spectroscopy (OES) data, which represent plasma process information obtained from faulty process scenario with intentional modification of the gas flow rate in a semiconductor fabrication process. We achieved a modeling prediction accuracy of modeling of 94.69% with selected SVID and OES and an accuracy of 93.68% with OES data alone. To conclude, the possibility of using an AIS in the field of semiconductor process decision making is proposed.

Semiconductor manufacturing involves a complex sequence of unit processes, where even a minor error can disrupt the entire production chain. Present-day manufacturing setups rely on continuous data monitoring of equipment health, wafer measurements, and inspections to identify any abnormalities that could impact the quality and performance of the final chip. The primary aim is fault detection and classification (FDC), for which a range of techniques including statistical analysis and machine learning algorithms are commonly employed. In this study, we introduce an innovative approach utilizing an artificial immune system (AIS), inspired by biological mechanisms, for FDC in semiconductor equipment. The main culprits behind process failures are shifts caused by the aging of parts and modules over time. Our methodology integrates state variable identification (SVID) data, reflecting current equipment conditions, and optical emission spectroscopy (OES) data, capturing plasma process information under faulty scenarios induced by deliberate gas flow rate adjustments in semiconductor fabrication. Our results demonstrate a modeling prediction accuracy of 94.69% when incorporating selected SVID and OES data, and 93.68% accuracy using OES data alone. In conclusion, we suggest the potential application of AIS in semiconductor process decision-making, offering promising avenues for enhancing fault detection in semiconductor equipment.

Functional data characterize the quality or reliability performance of many manufacturing processes. As can be seen in the literature, such data are informative in process monitoring and control for nanomachining, for ultra-thin semiconductor fabrication, and for antenna, steel-stamping, or chemical manufacturing processes. Many functional data in manufacturing applications show complicated transient patterns such as peaks representing important process characteristics. Wavelet transforms are popular in the computing and engineering fields for handling these types of complicated functional data. This article develops a wavelet-based statistical process control (SPC) procedure for detecting ‘out-of-control’ events that signal process abnormalities. Simulation-based evaluations of average run length indicate that our new procedure performs better than extensions from well-known methods in the literature. More importantly, unlike recent SPC research on linear profile data for monitoring global changes of data patterns, our methods focus on local changes in data segments. In contrast to most of the SPC procedures developed for detecting a known type of process change, our idea of updating the selected parameters adaptively can handle many types of process changes whether known or unknown. Finally, due to the data-reduction efficiency of wavelet thresholding, our procedure can deal effectively with large data sets.

Fault Detection and Diagnosis (FDD) in complex industrial systems requires methods that can handle uncertain operating conditions, soft thresholds, evolving sensor behavior, and increasing volumes of heterogeneous data. Traditional model-based or rule-driven approaches offer interpretability but lack adaptability, while purely data-driven and Large Language Model (LLM)-based methods often struggle with consistency, traceability, and causal grounding. Dynamic Master Logic (DML) provides a causal and temporal reasoning structure with fuzzy rules that capture gradual drift, soft limits, and asynchronous sensor signals while preserving traceability and deterministic evidence propagation. Building on this foundation, this paper presents a DML-LLM hybrid architecture that integrates targeted LLM inference to interpret unstructured information such as logs, notes, or retrieved documents under controlled prompts that maintain domain constraints. The combined system integrates Bayesian updating, deterministic routing, and semantic interpretation into a unified FDD pipeline. In a semiconductor manufacturing case study, the proposed framework reduced time to detection (TTD) from 7.4 h to 1.2 h and improved the F1 score from 0.59 to 0.83 when compared with conventional Statistical Process Control (SPC) and Fault Detection and Classification (FDC) workflows. Provenance completeness increased from 18% to 96%, while engineer triage time was reduced from 72 min to 18 min per event. These results demonstrate that the hybrid framework provides a scalable and explainable approach to anomaly detection and fault diagnosis in sensor-rich industrial environments.

Multizone pressure chemical mechanical planarization (CMP), which is regarded as irreplaceable in semiconductor manufacturing, is highly dependent on a process control strategy. However, predictive models with strict rules for advanced CMP multizone pressure process control have rarely been examined, even with the development of machine learning applications in semiconductor fabrication. Establishing a precise data-driven model that takes the fabrication conditions into consideration is urgent for controlling the CMP process. A novel structure for the CMP material removal rate model consisting of two parts is proposed and validated according to split pressure experiments and feature importance learning by this article. The first part was devoted to the prediction of the polishing time and time dependent problem with a deep neural network. The other part focused on the mutual effect of the pressure distribution on the multizone material removal rate using deep neural network with fully connected layers. To improve the prediction model of the first part, this study devises an advanced predictive model that combines a bidirectional gated recurrent unit network with a multi-head self-attention module and a one-dimensional neural network. This deep network extracts time-varying factors from industrial dataset and establish a nonlinear mapping function from CMP process features to material removal rate with no pressure variables. Both part of this structure is tested and evaluated for the feasibility of the predictive model using dataset from industrial fabrication.

Integration of advanced process control and full-scale dynamic simulation

As manufacturing quality has become a decisive factor in global market competition, statistical quality techniques such as Statistical Process Control (SPC) are widely used in industry. With advances in information, sensing, and data collection technology, large volumes of data are routinely available in processes employing Automatic Process Control (APC) and Engineering Process Control (EPC). Although there is a growing need for SPC monitoring in these feedback-controlled environments, an effective implementation scheme is still lacking. This research provides a monitoring method, termed the dynamic T1 chart that improves the detection of assignable causes in feedback-controlled processes.

This article considers the statistical adaptive process control for two dependent process steps. We construct an adaptive sampling interval Z X control chart to monitor the quality variable produced by the first process step, and use the adaptive sampling interval Z e control chart to monitor the specific quality variable produced by the second process step. By using the proposed adaptive sampling interval control charts, we can quickly detect and distinguish which process step is out of control. The performance of the proposed adaptive sampling interval control charts is measured by the adjusted average time to signal (AATS), which was derived by a Markov chain approach, for an out-of-control process. An empirical automobile braking system example shows the application and the performance of the proposed adaptive sampling control charts in detecting shifts in process means. Some numerical results obtained demonstrated that the performance of the proposed adaptive sampling cause-selecting control charts outperforms the fixed sampling interval cause-selecting control charts.

The aim of this research is to expand a framework that integrates two important concepts: statistical process control (SPC) and engineering process control (EPC). Most of the literature researches on integrated SPC/EPC systems are focused into continuous process mainly with algorithmic SPC. The integrated SPC/EPC systems in batch process control have not received the same degree of attention. This paper is a first of its kind in integrated SPC/EPC systems that applied in batch process. The proposed SPC/EPC integration is performed continually in active SPC for the batch making advance and in run-to-run (RTR) control action between batches. As a validation step, the proposed approach is applied to an industrial batch alkyd polymerisation reactor. Through this case study application, process engineers at the company are now able to use a valuable decision making tool when the production process is affected by certain disruptions, with obvious consequences on product quality, productivity and competitiveness.

Reliability of analytical systems: use of control charts, time series models and recurrent neural networks (RNN)

An advanced multivariable in-line process control system, which combines traditional statistical process control (SPC) with feedback control, has been applied to the CVD tungsten process on an Applied Materials reactor. The goal of the model-based controller is to compensate for shifts in the process and maintain the wafer-state responses on target. The controller employs measurements made on test wafers to track the process behavior. This is accomplished by using model-based SPC, which compares the measurements with predictions obtained from process models. The process models relate the equipment settings to the wafer-state responses of interest. For CVD tungsten, a physically-based modeling approach was employed based on the reaction rate for the H/sub 2/ reduction of WF/sub 6/. The Arrhenius relationship for the kinetic model was linearized so that empirical modeling techniques could be applied. Statistically valid models were derived for deposition rate, film stress, and bulk resistivity using stepwise least-squares regression. On detecting a statistically significant shift in the process, the controller calculates adjustments to the settings to bring the process responses back on target. To achieve this, two additional test wafers are processed at slightly different settings than the current recipe. This local experiment allows the models to be updated to reflect the current process state. The model updates are expressed as multiplicative or additive changes in the process inputs and a change in the model constant. This approach for adaptive control also provides a diagnostic capability regarding the cause of the process shift. The adapted models are used by an optimizer to compute new settings to bring the responses back to target. The optimizer is capable of incrementally entering controllables into the strategy, reflecting the degree to which the engineer desires to manipulate each setting. The capability of the controller to compensate for induced shifts in the CVD tungsten process is demonstrated. Targets for film bulk resistivity and deposition rate were maintained while satisfying constraints on film stress and WF/sub 6/ conversion efficiency. The ability of the controller to update process models during routine operation is also investigated. The tuned process models better predict the process behavior over time compared to the untuned models and lead to improved process capability.

Advanced process control (APC) refers to a broad range of techniques related processment and its application ground level and technologies implemented within industrial processment in control systems modification criteria. Advanced process controls are usually deployed optionally which is observed and additionally to basic process controls and measurement system. Basic process controls are designed and organized built with the process itself which is discussed here; to facilitate basic operation system and method area, control and automation requirements for this purpose related (APC). Advanced process controls are typically added subsequently by term, often over the course of many years which is mention here, to address particular performance economic area development opportunities in theprocessment system idea sector. (APC) Advance Process Control.

Tremendous amounts of equipment parameters and sensor values were generated during modern semiconductor wafer manufacturing. These process data were utilized for early diagnosis of anomalies to prevent subsequent yield loss. However, process data from different steps and machines were highly customized that it is of great difficulties for traditional fault detection and classification (FDC) analysis to find a universal model to identify process excursions. In this paper, we present a neural network method with deep convolutional variational autoencoder structures which used reconstruction error as abnormal scores. This method exhibited a more reliable detection precision than FDC method and traditional anomaly detection algorithms. Furthermore, the reconstruction error of different sensors could be used to identify the root cause of abnormal processes. It was found that similarly structured networks could be applied to different processing steps, which enable this method to be accepted as a standard method to process data.

Advanced Process Control Modeling from an Advanced Planning and Scheduling Perspective