基于局部与非局部线性判别分析和高斯混合模型动态集成的晶圆表面缺陷探测与识别

在复杂的半导体制造过程中,晶圆生产经过薄膜沉积、蚀刻、抛光等多项复杂的工序,制造过程中的异常波动都可能导致晶圆缺陷产生.晶圆表面的缺陷模式通常反映了半导体制造过程的各种异常问题,生产线上通过探测和识别晶圆表面缺陷,可及时判断制造过程故障源并进行在线调整,降低晶圆成品率损失.本文提出了基于一种流形学习算法与高斯混合模型动态集成的晶圆表面缺陷在线探测与识别模型.首先该模型开发了一种新型流形学习算法——局部与非局部线性判别分析法(Local and nonlocal linear discriminant analysis, LNLDA),通过融合数据局部/非局部信息以及局部/非局部惩罚信息,有效地提取高维晶圆特征数据的内在流形结构信息,以最大化数据不同簇样本的低维映射距离,保持特征数据中相同簇的低维几何结构.针对线上晶圆缺陷产生的随机性和复杂性,该模型对每种晶圆缺陷模式构建相应的高斯混合模型(Gaussian mixture model, GMM),提出了基于高斯混合模型动态集成的晶圆缺陷在线探测与识别方法.本文提出的模型成功地应用到实际半导体制造过程的晶圆表面缺陷在线探测与识别,在WM-811K晶圆数据库的实验结果验证了该模型的有效性与实用性.     

Artificial intelligence-based sampling planning system for dynamic manufacturing process

The dynamic sampling instead of static sampling can more effectively utilize the inspection capacity for quicker excursion detection and increase the throughput of inspection machines without affecting the quality of inspection, so that achieve cycle time reduction. Accordingly, many researchers and semiconductor fabs start investigating the impacts of using dynamic sampling and there is currently a growing need for the dynamic sampling strategies in today's highly competitive semiconductor industry. Meanwhile, the use of artificial intelligence (AI) for knowledge discovery has become more common in industrial and manufacturing process control systems and recent advances in technology, particularly in networking, and information processing, have made the implementation of dynamic process sampling feasible. In this paper the optimal dynamic sampling method and the associated decision process based on AI technique are proposed and the effectiveness of them is validated through actual data sets collected from a semiconductor fabrication line. Finally, we present an AI-based dynamic sampling planning system incorporated the proposed methodology, which possesses four sub-components: wafer bin map (WBM) data mart, optimal sampling method generator (OSMG), sampling knowledge, and sampling adaptation monitor. Our research results provide an effective solution to implement a successful dynamic process sampling.     

An empirical study of design-of-experiment data mining for yield-loss diagnosis for semiconductor manufacturing

To maintain competitive advantages, semiconductor industry has strived for continuous technology migrations and quick response to yield excursion. As wafer fabrication has been increasingly complicated in nano technologies, many factors including recipe, process, tool, and chamber with the multicollinearity affect the yield that are hard to detect and interpret. Although design of experiment (DOE) is a cost effective approach to consider multiple factors simultaneously, it is difficult to follow the design to conduct experiments in real settings. Alternatively, data mining has been widely applied to extract potential useful patterns for manufacturing intelligence. However, because hundreds of factors must be considered simultaneously to accurately characterize the yield performance of newly released technology and tools for diagnosis, data mining requires tremendous time for analysis and often generates too many patterns that are hard to be interpreted by domain experts. To address the needs in real settings, this study aims to develop a retrospective DOE data mining that matches potential designs with a huge amount of data automatically collected in semiconductor manufacturing to enable effective and meaningful knowledge extraction from the data. DOE can detect high-order interactions and show how interconnected factors respond to a wide range of values. To validate the proposed approach, an empirical study was conducted in a semiconductor manufacturing company in Taiwan and the results demonstrated its practical viability.     

Novel method for detection of mixed-type defect patterns in wafer maps based on a single shot detector algorithm

Journal of Intelligent Manufacturing - In semiconductor manufacturing, detecting defect patterns is important because they are directly related to the root causes of failures in the wafer process....     

晶圆表面缺陷模式的在线探测与自适应识别研究

晶圆表面的缺陷通常反映了半导体制造过程存在的异常问题,通过探测与识别晶圆表面缺陷模式,可及时诊断故障源并进行在线调整。提出了一种晶圆表面缺陷模式的在线探测与自适应识别模型。首先该模型对晶圆表面的缺陷模式进行特征提取,基于特征集对每种晶圆模式构建相应的隐马尔科夫模型（Hidden Markov Model,HMM）,并提出基于HMM动态集成的晶圆缺陷在线探测与识别方法。提出的模型成功应用于WM-811K数据库的晶圆缺陷检测与识别中,实验结果充分证明了该模型的有效性与实用性。     

Defect spatial pattern recognition using a hybrid SOM–SVM approach in semiconductor manufacturing

As manufacturing geometries continue to shrink and circuit performance increases, fast fault detection and semiconductor yield improvement is of increasing concern. Circuits must be controlled to reduce parametric yield loss, and the resulting circuits tested to guarantee that they meet specifications. In this paper, a hybrid approach that integrates the Self-Organizing Map and Support Vector Machine for wafer bin map classification is proposed. The log odds ratio test is employed as a spatial clustering measurement preprocessor to distinguish between the systematic and random wafer bin map distribution. After the smoothing step is performed on the wafer bin map, features such as co-occurrence matrix and moment invariants are extracted. The wafer bin maps are then clustered with the Self-Organizing Map using the aforementioned features. The Support Vector Machine is then applied to classify the wafer bin maps to identify the manufacturing defects. The proposed method can transform a large number of wafer bin maps into a small group of specific failure patterns and thus shorten the time and scope for troubleshooting to yield improvement. Real data on over 3000 wafers were applied to the proposed approach. The experimental results show that our approach can obtain over 90% classification accuracy and outperform back-propagation neural network.     

Designing a new model of distributed quality control for sub-assemble products based on the intelligent web information system

The quality control of sub-assemble products (SAP) in a distributed manufacturing shop (DMS) becomes crucial and complicated when the production of SAPs involves a variety of production technology. In this case, traditional statistical process control methods are not sufficient to control such manufacturing system. Here, we design an intelligent web information system, where quality data are collected from DMS and stored in the central database. The processes of manufacturing SAPs in DMS are then controlled by clustering homogenous SAPs using the quality control of SAP in DMS (QCSD) and process smoothness factor based SAP predefined clustering (PSFSPC) algorithms, respectively. A prototype system called an intelligent web information system quality control (IWIS-QC) has been developed to trace the quality profiles of SAPs. Finally, a case study has been presented to illustrate and validate the proposed approach.     

Optimization of Wafer Exposure Patterns Using a Two-Dimensional Cutting Algorithm

The semiconductor industry plays an integral role in Taiwan's manufacturing sector. Although defect reduction has received considerable attention to improve the yield rate, the problem of optimizing wafer exposure patterns has seldom been addressed. This study formulates the wafer exposure-patterning problem into a cutting and packing problem by adopting an innovative approach. We developed a two-dimensional cutting algorithm to maximize the number of dies that can be produced from a wafer to increase the gross die yield. The proposed algorithm is successfully implemented in a wafer fabrication factory. Experimental results validate the effectiveness of the proposed algorithm.     

An Intelligent Online Monitoring and Diagnostic System for Manufacturing Automation

Condition monitoring and fault diagnosis in modern manufacturing automation is of great practical significance. It improves quality and productivity, and prevents damage to machinery. In general, this practice consists of two parts: 1)extracting appropriate features from sensor signals and 2)recognizing possible faulty patterns from the features. Through introducing the concept of marginal energy in signal processing, a new feature representation is developed in this paper. In order to cope with the complex manufacturing operations, three approaches are proposed to develop a feasible system for online applications. This paper develops intelligent learning algorithms using hidden Markov models and the newly developed support vector techniques to model manufacturing operations. The algorithms have been coded in modular architecture and hierarchical architecture for the recognition of multiple faulty conditions. We define a novel similarity measure criterion for the comparison of signal patterns which will be incorporated into a novel condition monitoring system. The sensor-based intelligent system has been implemented in stamping operations as an example. We demonstrate that the proposed method is substantially more effective than the previous approaches. Its unique features benefit various real-world manufacturing automation engineering, and it has great potential for shop floor applications.     

Evolving CBR and data segmentation by SOM for flow time prediction in semiconductor manufacturing factory

Flow time of semiconductor manufacturing factory is highly related to the shop floor status; however, the processes are highly complicated and involve more than 100 production steps. Therefore, a simulation model with the production process of a real wafer fab located in Hsin-Chu Science-based Park of Taiwan is built for further studying of the relationship between the flow time and the various input variables. In this research, a hybrid approach by combining Self-Organizing Map (SOM) and Case-Based Reasoning (CBR) for flow time prediction in semiconductor manufacturing factory is developed. And Genetic Algorithm (GA) is applied to fine-tune the weights of features in the CBR model. The flow time and related shop floor status are collected and fed into the SOM for clustering. Then, a corresponding SGA-CBR method is selected and applied for flow time prediction. Finally, using the simulated data, the effectiveness of the proposed method (SGA-CBR) is shown by comparing with other approaches.     

Multiple-objective scheduling and real-time dispatching for the semiconductor manufacturing system

Implementing efficient scheduling and dispatching policies is a critical means to gain competitiveness for modern semiconductor manufacturing systems. In contemporary global market, a successful semiconductor manufacturer has to excel in multiple performance indices, consequently qualified scheduling approaches should provide efficient and holistic management of wafer products, information and manufacturing resources and make adaptive decisions based on real-time processing status to reach an overall optimized system performance. To cope with this challenge, a timed extended object-oriented Petri nets (EOPNs) based multiple-objective scheduling and real-time dispatching approach is proposed in this paper. Four performance objectives pursued by semiconductor manufacturers are integrated into a priority-ranking algorithm that serves as the initial scheduling guidance, and then all wafer lots will be dynamically dispatched by the hybrid real-time dispatching control system. A set of simulation experiments validate the proposed multiple-objective scheduling and real-time dispatching algorithm may achieve satisfactory performances.     

Hierarchical indices to detect equipment condition changes with high dimensional data for semiconductor manufacturing

During semiconductor manufacturing process, massive and various types of interrelated equipment data are automatically collected for fault detection and classification. Indeed, unusual wafer measurements may reflect a wafer defect or a change in equipment conditions. Early detection of equipment condition changes assists the engineer with efficient maintenance. This study aims to develop hierarchical indices for equipment monitoring. For efficiency, only the highest level index is used for real-time monitoring. Once the index decreases, the engineers can use the drilled down indices to identify potential root causes. For validation, the proposed approach was tested in a leading semiconductor foundry in Taiwan. The results have shown that the proposed approach and associated indices can detect equipment condition changes after preventive maintenance efficiently and effectively.     

A combined dispatching criteria approach to scheduling semiconductor manufacturing systems

The scheduling problem of semiconductor manufacturing systems has multiple responses of interest. The objective is to simultaneously optimize these different responses or to find the best-compromised solution. Most previous research in the area of semiconductor manufacturing systems has focused on optimizing a single performance measure. Dabbas and Fowler proposed a modified dispatching approach that combines multiple dispatching criteria into a single rule with the objective of simultaneously optimizing multiple objectives. In this paper, we validate their proposed approach using two different fab models at different levels of complexity: a hypothetical six stage-five machines Mini-Fab model and a full scale wafer fab model adapted from an actual Motorola wafer fab. We also discuss the actual implementation of the proposed dispatching algorithm into a scheduler for daily operation at a Motorola wafer fabrication facility. Results show an average 20% improvement for all responses when using the proposed dispatching approach.     

Constructing a yield model for integrated circuits based on a novel fuzzy variable of clustered defect pattern

As the wafer size increases, the clustering phenomenon of defects becomes significant. In addition to clustered defects, various clustering patterns also influence the wafer yield. In fact, the recognition of clustering pattern usually exists fuzziness. However, the wafer yield models in previous studies did not consider the fuzziness of clustering pattern belonging to which shape in recognition. Therefore, the objective of this study is to develop a new fuzzy variable of clustering pattern (FVCP) by using fuzzy logic control, and predict the wafer yield by using back-propagation neural network (BPNN) incorporating ant colony optimization (ACO). The proposed method utilizes defect counts, cluster index (CI), and FVCP as inputs for ACO-BPNN. A simulated study is utilized to demonstrate the effectiveness of the proposed model.     

The application of control chart for defects and defect clustering in IC manufacturing based on fuzzy theory

c-Chart was frequently used to monitor wafer defects during IC manufacturing. The clustering degree of defect on a wafer will increase along with the area of wafer gradually enlarging. The defect clustering causes the Poisson-based c-chart to exhibit many false alarms. Although several revised control charts have been developed to reduce the number of false alarms, those control charts still have some disadvantages in practical use. This study proposes a control chart that applies fuzzy theory and engineering experience to monitor wafer defects with the consideration of defect clustering. The proposed control chart is simpler and more rational than those revised c-charts. Finally, a case study of an IC company, owing to the HsinChu Scientific part at Taiwan, is used to demonstrate and verify the rationality and effectiveness.     

An online collaborative semiconductor yield forecasting system

Yield forecasting is a very important task to a semiconductor manufacturing factory which is a typical group-decision-making environment. Namely, many experts will gather to predict the yields of products collaboratively. To enhance both the precision and accuracy of collaborative semiconductor yield forecasting, an online expert system is constructed in this study. The collaborative semiconductor yield forecasting system adopts the client–server architecture, and therefore the necessity for all experts to gather at the same place is relaxed, which is especially meaningful for a multiple-factory case. To demonstrate the applicability of the collaborative semiconductor yield forecasting system, an experimental system has been constructed and applied to two random-access-memory products in a real semiconductor manufacturing factory. Both the precision and accuracy of forecasting the yields of the two products were significantly improved. Besides, the collaborative semiconductor yield forecasting system was also considered as a convenient platform for the product engineers or quality control staff from different factories to share their opinions about the yield improvement process of a product being manufacturing with the same technology in multiple factories.     

Dynamic vehicle allocation control for automated material handling system in semiconductor manufacturing

The current study examines the dynamic vehicle allocation problems of the automated material handling system (AMHS) in semiconductor manufacturing. With the uncertainty involved in wafer lot movement, dynamically allocating vehicles to each intrabay is very difficult. The cycle time and overall tool productivity of the wafer lots are affected when a vehicle takes too long to arrive. In the current study, a Markov decision model is developed to study the vehicle allocation control problem in the AMHS. The objective is to minimize the sum of the expected long-run average transport job waiting cost. An interesting exhaustive structure in the optimal vehicle allocation control is found in accordance with the Markov decision model. Based on this exhaustive structure, an efficient algorithm is then developed to solve the vehicle allocation control problem numerically. The performance of the proposed method is verified by a simulation study. Compared with other methods, the proposed method can significantly reduce the waiting cost of wafer lots for AMHS vehicle transportation.     

晶圆表面缺陷模式识别的二维主成分分析卷积自编码器

由于半导体制造过程的高度复杂性和动态性,各种过程故障通常导致晶圆表面出现各种缺陷模式.为了有效地识别晶圆表面缺陷模式从而及时地诊断和控制故障源,提出一种深度神经网络模型--二维主成分分析卷积自编码器(two-dimensional principal component analysis-based convolutional autoencoder, PCACAE).首先,提出一种基于改进的二维主成分分析算法(conditional2DPCA,C2DPCA)的图像卷积核,形成PCACAE的第1个卷积层;其次,对卷积输出进行池化操作并卷积编码重构,构建一个卷积编码器,并提取其编码部分作为PCACAE的第2层卷积层的初始化权值,从而形成一个深度网络模型,实现晶圆图像的特征学习;最后, PCACAE网络进行训练微调得到最终网络模型.将PCACAE应用于WM-811K晶圆图像数据库并与其他算法进行对比测试,实验结果表明, PCACAE在晶圆表面缺陷识别上的性能优于其他经典的卷积神经网络模型(如GoogLeNet,DensNet等),从而验证了该方法的有效性与工业可应用性.     

基于Petri网的组合设备建模与调度综述

组合设备是半导体晶圆制造的核心装备, 其调度与控制优化是半导体制造领域极具挑战性的课题. Petri网因其强大的建模能力和简约的图形化表达优势, 被广泛地应用于组合设备的建模与调度. 对基于Petri网的组合设备建模与调度方法进行综述, 归纳总结了组合设备的结构类型、晶圆流模式、调度策略及Petri网建模方法, 并系统阐述组合设备的7类典型调度问题, 包括驻留时间约束、作业时间波动、晶圆重入加工、多品种晶圆加工、加工模块(Process module, PM)故障、PM清洗和组合设备群. 最后, 讨论了当前组合设备调度存在的挑战及后续可能的研究方向.     

Development of Wafer Warehouse Management System for Semiconductor Manufacturing Foundry Supply Chain

Efficiency of supply chains management mostly depends on the process coordination and information integration between the supply chain companies.The well-known integrated circuit design houses,the wafer fabrication industries, and the integrated circuit packaging/testing business has together formed a contiguous supply chain from materials to system in Taiwan during the past decades.Logistic management of the wafer hence becomes the key linkage in the semiconductor foundry supply chain.The objective of this paper is to develop the wafer warehouse management system for global wafer logistics.Current operations for wafer logistics management are firstly reviewed. The system requirements are analyzed by the model-driven business transformation approach.The business operation model and the platform-independent solution architecture for the wafer logistics management are constructed.A prototype information system is also developed for validation.Results of this research can improve the effectiveness and efficiency in wafer logistics management for the semiconductor foundry supply chain.