Search

The annual output value of Taiwan’s wafer foundry and IC packaging test ranks first worldwide. The electronics industry has a complete industrial ecological chain in the supply chain system of global information and communication technology. In order to increase market competitiveness and increase the operational flexibility, manufacturing process outsourcing has become a trend in the business model of the electronics industry. The quality of the outsourcing affects the quality and function of the final product, the outsourcer selection is crucial to ensure the quality of the final product. This article takes the wire bonding process as an example, and proposes a fuzzy selection model of quality-based IC packaging process outsourcers. Then, this article selects the six standard deviation indicator that can fully reflect the process yield and quality level as the evaluation tool. Since the indicator contains unknown parameter, in order to improve the accuracy of estimation and overcome the uncertainty of measurement data, this paper applies mathematical programming to derive the confidence interval of six sigma quality index, and proposes a confidence-interval-based fuzzy testing meth as a green outsourcer selection tool to reduce the risk of misjudgment due to sampling errors and improve the accuracy of the selection. [ABSTRACT FROM AUTHOR]

In the semiconductor industry, the testing section has always played an important role. The testing section often requires engineers to judge the defect, which wastes a lot of time and cost. The accurate classification can provide useful information for engineers through neural networks. In this paper, we present a method for wafer map data augmentation and defect classification. Data augmentation is based on CNN encoder-decoder and the classification is based on depthwise separable convolutions. There are two datasets used, one is open dataset WM-811K and the other is built with a Taiwan company. We train two models with mobilenetV1 and V2 for the different datasets. The light-weight deep convolution can reduce model parameters and calculations, which is very efficient for the testing house with large production volumes. On two different data sets, our proposed method can reduce the number of parameters by 30% and 95%, and reduce the amount of calculation by 75% and 25%, respectively. The test accuracy of the first dataset is 93.95%. The second dataset test accuracy is 87.04%. After the data augmentation, accuracy is increased to 97.01% and 95.09%, respectively. [ABSTRACT FROM AUTHOR]

In the semiconductor manufacturing industry the development of single index assessing the equipment health condition is urgent, and thus the dashboard can be used for monitoring thousands of equipment. This paper proposes a novel equipment health index (EHI) framework and a Hotelling T-squared index (HTI) to monitor equipment and support preventive maintenance in real time in the semiconductor manufacturing industry. The EHI framework consists of data preprocessing, statistical process control, analytic hierarchy process, and a comprehensive single index representing the health condition of equipment. It considers different types of the univariate control charts for each status variable identification (SVID), and aggregates the scores corresponding to the control charts throughout the hierarchical structure. The HTI considers the correlation among the selected SVIDs and builds a multivariate index by using Hotelling T-squared statistic. An empirical study of Taiwan’s leading semiconductor assembly manufacturer finds that both the EHI and the HTI supported monitoring thousands of pieces of equipment in real time. In practice, firms can rapidly troubleshoot the root causes of failures by the decomposition of EHI. [ABSTRACT FROM AUTHOR]

Fault detection and classification has been employed to enhance yield and product quality for smart semiconductor manufacturing. For early detection of abnormal events that cause defects, the status variables identification data collected by the sensors embedded in advanced machines can be analyzed to derive the actions for advanced process control and advanced equipment control. However, the validity and effectiveness of fault detection and classification technologies may highly depend on domain knowledge and experience of the process engineers who should redefine the monitoring rules quickly when new process excursion occurred especially when ramping up new technologies and products. Motivated by realistic needs, this study aims to propose a novel strategy to empower intelligent fault detection and classification that employed convolutional neural network to analyze the feature SVID data and determine the conditions of the wafers, while shorten the cycle time for self-learning from domain knowledge and redefining new monitoring rules for fault classification in real time. This approach is validated with an empirical study in a leading semiconductor manufacturing company in Taiwan. The results have demonstrated the practical viability of the developed solution. [ABSTRACT FROM AUTHOR]