Search Results (13)

Semiconductor wafer manufacturing is one of the most complex and data-intensive industrial processes, comprising 500–1000 tightly interdependent steps, each requiring nanometer-level precision. As device nodes approach 3 nm and beyond, even minor deviations in parameters such as oxide thickness or critical dimensions can lead to catastrophic yield loss, challenging traditional physics-based control methods. In response, the industry has increasingly adopted regression analysis and predictive modeling as essential analytical frameworks. Classical regression, long used to support design of experiments (DOE), process optimization, and yield analysis, has evolved to enable multivariate modeling, virtual metrology, and fault detection. Predictive modeling extends these capabilities through machine learning and AI, leveraging massive sensor and metrology data streams for real-time process monitoring, yield forecasting, and predictive maintenance. These data-driven tools are now tightly integrated into advanced process control (APC), digital twins, and automated decision-making systems, transforming fabs into agile, intelligent manufacturing environments. This review synthesizes foundational and emerging methods, industry applications, and case studies, emphasizing their role in advancing Industry 4.0 initiatives. Future directions include hybrid physics–ML models, explainable AI, and autonomous manufacturing. Together, regression and predictive modeling provide semiconductor fabs with a robust ecosystem for optimizing performance, minimizing costs, and accelerating innovation in an increasingly competitive, high-stakes industry. Full article

Faults in the wafer transfer robots (WTRs) used in semiconductor manufacturing processes can significantly affect productivity. This study defines high-risk components such as bearing motors, ball screws, timing belts, robot hands, and end effectors, and generates fault data for each component based on Fluke’s law. A stacking classifier was applied for fault prediction and severity classification, and logistic regression was used to identify fault components. Additionally, to analyze the frequency bands affecting each failed component and assess the severity of faults involving two mixed components, a hybrid explainable artificial intelligence (XAI) model combining Shapley additive explanations (SHAP) and local interpretable model-agnostic explanations (LIME) was employed to inform the user about the component causing the fault. This approach demonstrated a high prediction accuracy of 95%, and its integration into real-time monitoring systems is expected to reduce maintenance costs, decrease equipment downtime, and ultimately improve productivity. Full article

Amid the ongoing emphasis on reducing manufacturing costs and enhancing productivity, one of the crucial objectives when manufacturing is to maintain process tools in optimal operating conditions. With advancements in sensing technologies, large amounts of data are collected during manufacturing processes, and the challenge today is to utilize these massive data efficiently. Some of these data are used for fault detection and classification (FDC) to evaluate the general condition of production machinery. The distinctive characteristics of semiconductor manufacturing, such as interdependent parameters, fluctuating behaviors over time, and frequently changing operating conditions, pose a major challenge in identifying defective wafers during the manufacturing process. To address this challenge, a multivariate fault detection method based on a 1D ResNet algorithm is introduced in this study. The aim is to identify anomalous wafers by analyzing the raw time-series data collected from multiple sensors throughout the semiconductor manufacturing process. To achieve this objective, a set of features is chosen from specified tools in the process chain to characterize the status of the wafers. Tests on the available data confirm that the gradient vanishing problem faced by very deep networks starts to occur with the plain 1D Convolutional Neural Network (CNN)-based method when the size of the network is deeper than 11 layers. To address this, a 1D Residual Network (ResNet)-based method is used. The experimental results show that the proposed method works more effectively and accurately compared to techniques using a plain 1D CNN and can thus be used for detecting abnormal wafers in the semiconductor manufacturing industry. Full article

The semiconductor industry is one of the most technology-evolving and capital-intensive market sectors. Effective inspection and metrology are necessary to improve product yield, increase product quality and reduce costs. In recent years, many types of semiconductor manufacturing equipments have been equipped with sensors to facilitate real-time monitoring of the production processes. These production-state and equipment-state sensor data provide an opportunity to practice machine-learning technologies in various domains, such as anomaly/fault detection, maintenance scheduling, quality prediction, etc. In this work, we focus on the soft-sensing regression problem in metrology systems, which uses sensor data collected during wafer processing steps to predict impending inspection measurements that used to be measured in wafer inspection and metrology systems. We proposed a regressor based on Long Short-term Memory network and devised two distinct loss functions for the purpose of the training model. Although the assessment of our prediction errors by engineers is subjective, a novel piece-wise evaluation metric was introduced to evaluate model accuracy in a mathematical way. Our experimental results showcased that the proposed model is capable of achieving both accurate and early prediction across various types of inspections in complicated manufacturing processes. Full article

To minimize wafer yield losses by misprocessing during semiconductor manufacturing, faster and more accurate fault detection during the plasma process are desired to increase production yields. Process faults can be caused by abnormal equipment conditions, and the performance drifts of the parts or components of complicated semiconductor fabrication equipment are some of the most unnoticed factors that eventually change the plasma conditions. In this work, we propose improved stability and accuracy of process fault detection using optical emission spectroscopy (OES) data. Under a controlled experimental setup of arbitrarily induced fault scenarios, the extended isolation forest (EIF) approach was used to detect anomalies in OES data compared with the conventional isolation forest method in terms of accuracy and speed. We also used the OES data to generate features related to electron temperature and found that using the electron temperature features together with equipment status variable identification data (SVID) and OES data improved the prediction accuracy of process/equipment fault detection by a maximum of 0.84%. Full article

The production process of a wafer in the semiconductor industry consists of several phases such as a diffusion and associated defectivity test, parametric test, electrical wafer sort test, assembly and associated defectivity tests, final test, and burn-in. Among these, the fault detection phase is critical to maintain the low number and the impact of anomalies that eventually result in a yield loss. The understanding and discovery of the causes of yield detractors is a complex procedure of root-cause analysis. Many parameters are tracked for fault detection, including pressure, voltage, power, or valve status. In the majority of the cases, a fault is due to a combination of two or more parameters, whose values apparently stay within the designed and checked control limits. In this work, we propose an ensembled anomaly detector which combines together univariate and multivariate analyses of the fault detection tracked parameters. The ensemble is based on three proposed and compared balancing strategies. The experimental phase is conducted on two real datasets that have been gathered in the semiconductor industry and made publicly available. The experimental validation, also conducted to compare our proposal with other traditional anomaly detection techniques, is promising in detecting anomalies retaining high recall with a low number of false alarms. Full article

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. Full article

Condition-based maintenance refers to the installation of permanent sensors on a structure/system. By means of early fault detection, severe damage can be avoided, allowing efficient timing of maintenance works and avoiding unnecessary inspections at the same time. These are the goals for structural health monitoring (SHM). The changes caused by incipient damage on raw data collected by sensors are quite small, and are usually contaminated by noise and varying environmental factors, so the algorithms used to extract information from sensor data need to focus on sensitive damage features. The developments of SHM techniques over the last 20 years have been more related to algorithm improvements than to sensor progress, which essentially have been maintained without major conceptual changes (with regards to accelerometers, piezoelectric wafers, and fiber optic sensors). The main different SHM systems (vibration methods, strain-based fiber optics methods, guided waves, acoustic emission, and nanoparticle-doped resins) are reviewed, and the main issues to be solved are identified. Reliability is the key question, and can only be demonstrated through a probability of detection (POD) analysis. Attention has only been paid to this issue over the last ten years, but now it is a growing trend. Simulation of the SHM system is needed in order to reduce the number of experiments. Full article

Micro-electronic devices are increasingly incorporating miniature multi-layered integrated architectures. However, the localization of faults in three-dimensional structure remains challenging. This study involved the experimental and numerical estimation of the depth of a thermally active heating source buried in multi-layered silicon wafer architecture by using both phase information from an infrared microscopy and finite element simulation. Infrared images were acquired and real-time processed by a lock-in method. It is well known that the lock-in method can increasingly improve detection performance by enhancing the spatial and thermal resolution of measurements. Operational principle of the lock-in method is discussed, and it is represented that phase shift of the thermal emission from a silicon wafer stacked heat source chip (SSHSC) specimen can provide good metrics for the depth of the heat source buried in SSHSCs. Depth was also estimated by analyzing the transient thermal responses using the coupled electro-thermal simulations. Furthermore, the effects of the volumetric heat source configuration mimicking the 3D through silicon via integration package were investigated. Both the infrared microscopic imaging with the lock-in method and FE simulation were potentially useful for 3D isolation of exothermic faults and their depth estimation for multi-layered structures, especially in packaged semiconductors. Full article

The use of piezoelectric wafer active sensors (PWAS) for structural health monitoring (SHM) purposes is state of the art for acousto-ultrasonic-based methods. For system reliability, detailed information about the PWAS itself is necessary. This paper gives an overview on frequent PWAS faults and presents the effects of these faults on the wave propagation, used for active acousto-ultrasonics-based SHM. The analysis of the wave field is based on velocity measurements using a laser Doppler vibrometer (LDV). New and established methods of PWAS inspection are explained in detail, listing advantages and disadvantages. The electro-mechanical impedance spectrum as basis for these methods is discussed for different sensor faults. This way this contribution focuses on a detailed analysis of PWAS and the need of their inspection for an increased reliability of SHM systems. Full article

In order to meet the requirements of cable fault detection, a new structure of piezoelectric accelerometer was designed and analyzed in detail. The structure was composed of a seismic mass, two sensitive beams, and two added beams. Then, simulations including the maximum stress, natural frequency, and output voltage were carried out. Moreover, comparisons with traditional structures of piezoelectric accelerometer were made. To verify which vibration mode is the dominant one on the acceleration and the space between the mass and glass, mode analysis and deflection analysis were carried out. Fabricated on an n-type single crystal silicon wafer, the sensor chips were wire-bonged to printed circuit boards (PCBs) and simply packaged for experiments. Finally, a vibration test was conducted. The results show that the proposed piezoelectric accelerometer has high sensitivity, low resonance frequency, and low transverse effect. Full article

Plasma process tools, which usually cost several millions of US dollars, are often used in the semiconductor fabrication etching process. If the plasma process is halted due to some process fault, the productivity will be reduced and the cost will increase. In order to maximize the product/wafer yield and tool productivity, a timely and effective fault process detection is required in a plasma reactor. The classification of fault events can help the users to quickly identify fault processes, and thus can save downtime of the plasma tool. In this work, optical emission spectroscopy (OES) is employed as the metrology sensor for in-situ process monitoring. Splitting into twelve different match rates by spectrum bands, the matching rate indicator in our previous work (Yang, R.; Chen, R.S. Sensors 2010, 10, 5703-5723) is used to detect the fault process. Based on the match data, a real-time classification of plasma faults is achieved by a novel method, developed in this study. Experiments were conducted to validate the novel fault classification. From the experimental results, we may conclude that the proposed method is feasible inasmuch that the overall accuracy rate of the classification for fault event shifts is 27 out of 28 or about 96.4% in success. Full article

The plasma process is often used in the fabrication of semiconductor wafers. However, due to the lack of real-time etching control, this may result in some unacceptable process performances and thus leads to significant waste and lower wafer yield. In order to maximize the product wafer yield, a timely and accurately process fault or abnormal detection in a plasma reactor is needed. Optical emission spectroscopy (OES) is one of the most frequently used metrologies in in-situ process monitoring. Even though OES has the advantage of non-invasiveness, it is required to provide a huge amount of information. As a result, the data analysis of OES becomes a big challenge. To accomplish real-time detection, this work employed the sigma matching method technique, which is the time series of OES full spectrum intensity. First, the response model of a healthy plasma spectrum was developed. Then, we defined a matching rate as an indictor for comparing the difference between the tested wafers response and the health sigma model. The experimental results showed that this proposal method can detect process faults in real-time, even in plasma etching tools. Full article