当前位置: X-MOL 学术ACM Trans. Intell. Syst. Technol. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Contextual Anomaly Detection in Solder Paste Inspection with Multi-Task Learning
ACM Transactions on Intelligent Systems and Technology ( IF 5 ) Pub Date : 2020-07-07 , DOI: 10.1145/3383261
Zimu Zheng 1 , Jie Pu 2 , Linghui Liu 2 , Dan Wang 3 , Xiangming Mei 2 , Sen Zhang 2 , Quanyu Dai 3
Affiliation  

In this article, we study solder paste inspection (SPI), an important stage that is used in the semiconductor manufacturing industry, where abnormal boards should be detected. A highly accurate SPI can substantially reduce human expert involvement, as well as reduce the waste in disposing of the boards in good condition. A key difference today is that because of increasing demand in board customization, the number of board types increases substantially and quantity of the boards produced in each type decreases. Thus, the previous approaches where a fine-tuned model is developed for each board type are no longer viable. Intrinsically, our problem is an anomaly detection problem. A major specialty in today’s SPI is that the target tasks for prediction cannot be fully pre-determined due to context changes during the solder paste printing stage. Our experiences show that a conventional approach to first define a set of tasks and train these tasks offline will lead to low accuracy. Here, we propose a novel multi-task approach, where the performance of all target tasks is ensured simultaneously. We note that the SPI process is streamlined and automatic, allowing the SPI time for only a few seconds. We propose a fast clustering algorithm that reuses existing models to avoid retraining and fine tune in the inference phase. We evaluate our approach using 3-month data collected from production lines. We show that we can reduce 81.28% of false alarms. This can translate to annual savings of $11.3 million.

中文翻译:

基于多任务学习的锡膏检测中的上下文异常检测

在本文中,我们研究了焊膏检测 (SPI),这是半导体制造业中使用的一个重要阶段,应该检测异常板。高度准确的 SPI 可以大大减少人工专家的参与,并减少在处理完好的电路板时的浪费。今天的一个关键区别是,由于对电路板定制的需求不断增加,电路板类型的数量大幅增加,而每种类型的电路板生产数量减少。因此,以前为每种电路板类型开发微调模型的方法不再可行。从本质上讲,我们的问题是异常检测问题。当今 SPI 的一个主要特点是,由于焊膏印刷阶段的环境变化,预测的目标任务无法完全预先确定。我们的经验表明,首先定义一组任务并离线训练这些任务的传统方法将导致准确性低。在这里,我们提出了一种新颖的多任务方法,可以同时确保所有目标任务的性能。我们注意到 SPI 过程是流线型和自动的,允许 SPI 时间只有几秒钟。我们提出了一种快速聚类算法,该算法重用现有模型,以避免在推理阶段进行重新训练和微调。我们使用从生产线收集的 3 个月数据来评估我们的方法。我们表明我们可以减少 81.28% 的误报。这可以转化为每年节省 1130 万美元。同时确保所有目标任务的执行。我们注意到 SPI 过程是流线型和自动的,允许 SPI 时间只有几秒钟。我们提出了一种快速聚类算法,该算法重用现有模型,以避免在推理阶段进行重新训练和微调。我们使用从生产线收集的 3 个月数据来评估我们的方法。我们表明我们可以减少 81.28% 的误报。这可以转化为每年节省 1130 万美元。同时确保所有目标任务的执行。我们注意到 SPI 过程是流线型和自动的,允许 SPI 时间只有几秒钟。我们提出了一种快速聚类算法,该算法重用现有模型,以避免在推理阶段进行重新训练和微调。我们使用从生产线收集的 3 个月数据来评估我们的方法。我们表明我们可以减少 81.28% 的误报。这可以转化为每年节省 1130 万美元。我们表明我们可以减少 81.28% 的误报。这可以转化为每年节省 1130 万美元。我们表明我们可以减少 81.28% 的误报。这可以转化为每年节省 1130 万美元。
更新日期:2020-07-07
down
wechat
bug