《電子技術(shù)應(yīng)用》
您所在的位置:首頁(yè) > EDA與制造 > 設(shè)計(jì)應(yīng)用 > 數(shù)字孿生在半導(dǎo)體測(cè)試設(shè)備中的實(shí)踐
數(shù)字孿生在半導(dǎo)體測(cè)試設(shè)備中的實(shí)踐
電子技術(shù)應(yīng)用
祁建華1,馬詩(shī)旻2,何桂玲2
1.上海華嶺集成電路技術(shù)股份有限公司;2.上?;饟纯萍加邢薰?/div>
摘要: 數(shù)字孿生作為智能制造關(guān)鍵技術(shù)在工廠極具實(shí)用價(jià)值。通過(guò)在半導(dǎo)體測(cè)試服務(wù)公司建立數(shù)字孿生系統(tǒng),收集了探針臺(tái)二十多萬(wàn)條報(bào)警數(shù)據(jù)。結(jié)果顯示,超過(guò)90%的報(bào)警已通過(guò)數(shù)字孿生系統(tǒng)實(shí)現(xiàn)了遠(yuǎn)程控制解決。為探索報(bào)警處理的最佳方式,進(jìn)一步優(yōu)化數(shù)字孿生模型,并在模型中重現(xiàn)報(bào)警,通過(guò)模擬不同工況,與傳統(tǒng)方式相比,人工操作數(shù)字孿生系統(tǒng)處理報(bào)警可使設(shè)備等待時(shí)間減少64%以上。若采用AI后臺(tái)自行操作數(shù)字孿生系統(tǒng),設(shè)備等待時(shí)間可繼續(xù)降低78%。這表明人工對(duì)設(shè)備報(bào)警的反應(yīng)時(shí)間是影響設(shè)備等待時(shí)間的關(guān)鍵因素,因此,深化使用數(shù)字孿生模型自動(dòng)化處理報(bào)警可顯著降低設(shè)備等待時(shí)間,提升設(shè)備綜合利用率。
中圖分類號(hào):TN711.5 文獻(xiàn)標(biāo)志碼:A DOI: 10.16157/j.issn.0258-7998.244981
中文引用格式: 祁建華,馬詩(shī)旻,何桂玲. 數(shù)字孿生在半導(dǎo)體測(cè)試設(shè)備中的實(shí)踐[J]. 電子技術(shù)應(yīng)用,2024,50(9):25-30.
英文引用格式: Qi Jianhua,Ma Shimin,He Guiling. The practice of digital twinship in semiconductor testing equipment[J]. Application of Electronic Technique,2024,50(9):25-30.
The practice of digital twinship in semiconductor testing equipment
Qi Jianhua1,Ma Shimin2,He Guiling2
1.Sino IC Technology Co., Ltd. ;2.Huozhi Technology Co., Ltd.
Abstract: Digital twin is an essential technology in smart manufacturing and has practical value in factories. At a semiconductor testing services company, a digital twin system was used to collect over 200 000 alarm data points from probe stations. The analysis showed that more than 90% of the alarms were resolved remotely using the digital twin system. To improve the digital twin model and alarm management, alarms were replicated within the model by simulating different scenarios. Operating the digital twin system manually reduced equipment downtime by over 64% compared to traditional methods. Additionally, using an AI backend to automatically operate the digital twin system further reduced equipment waiting time by 78%. This highlights the importance of personnel response time to equipment alarms in affecting equipment downtime. Therefore, enhancing the use of digital twin models for automated alarm handling can significantly reduce equipment downtime and improve overall equipment utilization.
Key words : digital twin;semiconductor testing equipment;remote control;alarm handling;automation

引言

我國(guó)在2015年部署了“中國(guó)制造2025”戰(zhàn)略,全面推進(jìn)實(shí)施制造強(qiáng)國(guó)戰(zhàn)略[1]。智能制造是“中國(guó)制造2025”的主攻方向,而數(shù)字孿生(Digital Twin)是智能制造迅速發(fā)展的關(guān)鍵技術(shù)之一[2-3]。

數(shù)字孿生技術(shù)通過(guò)為物理世界中的客觀對(duì)象建立數(shù)字模型,實(shí)現(xiàn)對(duì)物理世界實(shí)體對(duì)象的實(shí)時(shí)監(jiān)測(cè)、分析和優(yōu)化[4]。數(shù)字孿生可實(shí)現(xiàn)物理世界與信息世界的交互融合[5],被用于解決實(shí)際的工程問(wèn)題。

Gartner在2017年開始連續(xù)三年將數(shù)字孿生列為十大戰(zhàn)略科技發(fā)展趨勢(shì)[6-8]。中國(guó)科學(xué)技術(shù)協(xié)會(huì)在其發(fā)布的2020重大科學(xué)問(wèn)題和工程技術(shù)難題中將“如何建立虛擬孿生理論和技術(shù)基礎(chǔ)并開展示范應(yīng)用”列為前沿科學(xué)問(wèn)題之一[9]。

智能制造數(shù)字孿生應(yīng)用已被廣泛應(yīng)用于航空航天、電力、船舶等行業(yè)[10-12]。在半導(dǎo)體制造業(yè)的數(shù)字化轉(zhuǎn)型中,借助數(shù)字孿生技術(shù)可以對(duì)產(chǎn)線全流程進(jìn)行實(shí)時(shí)監(jiān)控和管理,有效提高企業(yè)的生產(chǎn)效率[13]。

集成電路生產(chǎn)線的設(shè)備投資占據(jù)總資本支出的80%(據(jù)SEMI估計(jì))[14],提高設(shè)備綜合利用率對(duì)于降低生產(chǎn)成本至關(guān)重要。通過(guò)IoT和邊緣計(jì)算,采集和上傳設(shè)備數(shù)據(jù)后,數(shù)字孿生技術(shù)可完成設(shè)備的可視化監(jiān)控管理,實(shí)現(xiàn)提質(zhì)增效的價(jià)值[15-16],數(shù)字孿生正成為制造業(yè)數(shù)字化轉(zhuǎn)型的核心驅(qū)動(dòng)力[17]。

數(shù)字孿生的定義在不同領(lǐng)域存在分歧[18]。陶飛教授提出了數(shù)字孿生成熟度模型分析[19],該模型將數(shù)字孿生的成熟度分為L(zhǎng)0至L5共六個(gè)等級(jí)。本文將探討半導(dǎo)體測(cè)試服務(wù)企業(yè)如何在這六個(gè)等級(jí)中,打通設(shè)備物理世界與虛擬世界的數(shù)據(jù)通道,并在數(shù)據(jù)上進(jìn)行模擬分析,嘗試進(jìn)行部分預(yù)測(cè)和優(yōu)化工作。


本文詳細(xì)內(nèi)容請(qǐng)下載:

http://ihrv.cn/resource/share/2000006137


作者信息:

祁建華1,馬詩(shī)旻2,何桂玲2

(1.上海華嶺集成電路技術(shù)股份有限公司,上海 201203;

2.上?;饟纯萍加邢薰荆虾?201402)


Magazine.Subscription.jpg

此內(nèi)容為AET網(wǎng)站原創(chuàng),未經(jīng)授權(quán)禁止轉(zhuǎn)載。