给咱们团队写个160字的团队推荐词，这是人力发来的改了一半的。。。：该团队核心能力突出，成果丰硕且协同价值显著。在核心业务建设上，成功构建工艺智能底座，打通“数据-分析-推荐”全链路实现数据资产化与模型标准化；在刻蚀与薄膜工艺中落地统一工具链，明确客户可交付方向；建立统一建模评估体系并完成跨工艺迁移验证，推动软件从单项目向平台化升级。跨事业部协同高效，与多个事业部实现无缝合作验证，通过平台工具固化专家经验形成可复用方法论。同时有力支撑6款平台研发及5个BU与标准化适配，完成代码流、config等多项标准化工作，落地RobotCBB薄膜刻蚀通用版开发，全方位为业务发展赋能。
华创软件 —星创先锋团队奖 工艺配方智能优化技术突破团队
让价值创造者实现更高
2
让价值 创造者实现更高让价值
创造者实现更高目录
1
团队名称及人员构成
2
团队主要业绩
3
工作的改革 /创新情况
4
对华创软件发展影响和贡献
5
跨体系工作协同情况
NAURA Confidential
6
价值观及关键事件
3
让价值 创造者实现更高让价值
创造者实现更高一、团队 名称及人员 构成
NAURA Confidential
数字智能部
N1 二刻
N8 薄膜
软件运营
平台架构
需求统筹
需求输出
需求输出
计划协助
算法研发
算法底座
算法设计
算法设计
刻蚀工艺
数据收集
跑片验证
节点协同
薄膜工艺
数据收集
跑片验证
跨部门协作
工艺配方智能优化技术突破团队 工艺配方智能优化技术突破团队
平台架构
•平台架构设计
•算法底座搭建
•数字智能部
•张迪、史恩可
算法研发
•核心模型研发
•标准化可复用
•数字智能部
•易志伟、李晓睿杨骁
刻蚀工艺
•预测模型评估
•推优跑片验证
•N1 二刻
•朱海云、于凡秦凯马俊胜
薄膜工艺
•优化目标选定
•推优跑片验证
•N8 薄膜
•吴桂龙、张嘉显王海艳
工艺配方智能优化项目组
数字智能部
平台研发
架构设计
技术选型
算法服务
算法实现
算法设计
模型研发
推优集成
工艺 BU
刻蚀
需求提供
工艺支持
验证测试
PVD
需求提供
工艺支持
验证测试
CVDCVDCVD
需求提供
工艺支持
验证测试
ALDALD
需求提供
工艺支持
验证测试
4
让价值 创造者实现更高让价值
创造者实现更高二、团队主要 业绩
NAURA Confidential
ConfidentialNAURA Confidential NAURA ConfidentialNAURA ConfidentialNAURA ConfidentialNAURA ConfidentialNAURA Confidential NAURA ConfidentialNAURA Confidential
软件体系 工艺智能化力建设
数据处理
特征工程
模型训练
模型评估
模型解释
参数推优
流程 标准化
①DOEDOE 设计
④工艺 模型构建
参数空间设计
②DOEDOE 跑片
⑤配方智能优化
③数据收集
⑥配方 推荐
贝叶 斯优化
帕累托前沿
回归预测
高斯过程
Bayesian Bayesian Bayesian Bayesian Bayesian Bayesian
Optimization OptimizationOptimization Optimization OptimizationOptimization Optimization
Regression RegressionRegressionRegressionRegressionRegression Regression
PredictionPredictionPredictionPredictionPrediction PredictionPrediction Prediction
ParameterParameterParameterParameterParameterParameterParameterParameterParameter
Space designSpace designSpace designSpace designSpace designSpace designSpace designSpace designSpace designSpace design Space design
国际厂商 工艺配方优化
基于批量工艺数据， 构建预测模型通过智能寻优算法推荐工艺配方参数
L* HF -CL CL 范式
A* Applied Pro Applied Pro Applied ProApplied ProApplied Pro
Naura 工艺配方智能优化平台
国际厂商 -工艺配方优化能力 工艺配方优化能力 工艺配方优化能力 工艺配方优化能力
数据可视化、模型 训练、工艺预测推优 训练、工艺预测推优 训练、工艺预测推优 训练、工艺预测推优
5
让价值 创造者实现更高让价值
创造者实现更高三、团队工作汇报 – 平台功能实现
DemoDemo Demo列表 -数据上传
数据分析 -分布探查
统计分析 -敏感度分析
数据分析 -变化趋势
平台功能
模型 管理 -模型训练
模型应用 -预测分析
模型应用 -工艺推优
统计分析 -相关性分析
模型应用 -响应面分析
6
让价值 创造者实现更高让价值
创造者实现更高三、团队工作汇报 – 工艺落地验证
NAURA Confidential
ConfidentialNAURA Confidential NAURA ConfidentialNAURA ConfidentialNAURA ConfidentialNAURA ConfidentialNAURA Confidential NAURA ConfidentialNAURA Confidential
工艺
PVD TTN PVD TTN PVD TTNPVD TTNPVD TTN
工艺目标
RsU ＜ 3%
Rs 26.5
机台型号
eVictor eVictor eVictor MX30MX30MX30
模型性能指标
数据 情况
RecipeRecipe Recipe Recipe50 条
工艺参数 7个
达成 指标
R² ≥0.940.94 0.94
趋势 一致性 100%
响应面 分析
经工艺程师验证：响应面分析 结果符合工艺预期
工艺
ALD ALD ALD TiN
工艺目标
THK THK THK ： 13 ű0.3Å
U% <% <% <5%
Ratio Ratio Ratio ：1.0% ±0.1%
机台型号
eVictor eVictor eVictor MX30MX30MX30
模型性能指标
数据 情况
Recipe Recipe Recipe Recipe 18 条
工艺参数 6个
达成指标
R² ≥0.97 ≥0.97 ≥0.97 ≥0.97
MAE≤ MAE≤ MAE≤ 2%
推荐 命中率 100 %
预测值 vs vs 实测值 R²≥0.97 R²≥0.97 R²≥0.97 R²≥0.97
推荐配方命中结果
量测 THKTHKTHK
量测 U%
量测 RatioRatioRatio Ratio
推荐 1
13.1705
2.0936
1.0117
推荐 2
12.9889
2.0356
1.0033
推荐 3
12.7629
2.2275
1.0150
100% 命中工艺窗口
工艺
PECVD NDC PECVD NDC PECVD NDC PECVD NDCPECVD NDC
工艺目标
THK THK THK ：500A ±15
U%< 3%
RI ：1.90 ±0.01
机台型号
LYRA LYRA LYRA OptimaOptima OptimaOptimaOptima
模型性能指标
数据 情况
RecipeRecipe Recipe Recipe237 条
补点 24 条 工艺 参数 7个
迁移 指标
R² ≥0.960.96 0.96
MAE≤ MAE≤ 2%
推荐 命中率 100 %
预测值 vs vs 实测值 R²≥0.96 R²≥0.96 R²≥0.96 R²≥0.96
推荐配方命中结果
新腔室少量跑片 ，模型 微调迁移推荐参数， 100% 命中工艺窗口
量测 THKTHKTHK
量测 U%
量测 RI
推荐 1
498.507
2.856
1.8929
推荐 2
497.945
2.809
1.893
工艺
AA ReverseAA ReverseAA ReverseAA ReverseAA Reverse AA Reverse AA ReverseAA Reverse
工艺目标
TCD≤ TCD≤ TCD≤ TCD≤ 5nm 5nm
DepthDepth Depth≤100A
SiN SWA≤ SWA≤ 3°
SOC SOC SOC SOC remain≤100A remain≤100A remain≤100A remain≤100A
机台型号
612E LB
模型性能指标
数据 情况
RecipeRecipe Recipe Recipe40 条
工艺参数 54 个
步骤 14 个
达成 指标
R² ≥0.80.8
趋势 一致性 ≥90% ≥90% ≥90% ≥90%
敏感度分析
经工艺程师验证： 工艺趋势分析一致性 ≥90%
刻蚀
薄膜
工艺
NCHMNCHM NCHM
工艺目标
TCDTCDTCD48nm 48nm
MCD 50nmMCD 50nmMCD 50nmMCD 50nmMCD 50nm MCD 50nm
BCDBCDBCD50nm 50nm
LER 1.2 … LER 1.2 … LER 1.2 … LER 1.2 … LER 1.2 …
机台型号
612E LB
模型性能指标
数据 情况
RecipeRecipe Recipe Recipe30 条
工艺参数 70 个
步骤 6个
达成 指标
R² ≥0.80.8
趋势 一致性 ≥90% ≥90% ≥90%
参数推优
工艺程师验证： SPEC SPEC 由命中 2个提升至 6个，整体 形貌得到优化
工艺
28 HK28 HK28 HK28 HK28 HK
工艺目标
BCD 32BCD 32BCD 32BCD 32BCD 32 -36
HKBCD HKBCD HKBCD HKBCD HKBCD 32 -38
MCD MCD MCD MCD 31 -34
SWA 88SWA 88 SWA 88 -89.5 … 89.5 …
机台型号
612E LB
模型性能指标
数据 情况
RecipeRecipe Recipe Recipe42 条
工艺参数 48 个
达成指标
R² ≥0.80.8
趋势 一致性 ≥90% ≥90% ≥90% ≥90%
趋势 分析
经工艺程师验证： 工艺趋势分析一致性 ≥90%
7
让价值 创造者实现更高让价值
创造者实现更高AA ReverseAA ReverseAA ReverseAA Reverse AA ReverseAA Reverse
NCHM
28
High -K
三、团队工作创新 – 智能 模型构建
Level3 Level3 ：多工艺 跨腔室
层 次模型：工艺、腔室特征层 次模型：工艺、腔室特征层
次模型：工艺、腔室特征Level2 Level2 ：单工艺 跨腔室
迁移模型 ：基础+腔室偏移学习 腔室偏移学习 腔室偏移学习
Level1 Level1 ：单工艺 单腔室
基础模型：自适应邻域分析 基础模型：自适应邻域分析 +趋势学习 趋势学习
趋势继承
结构共享
智能优化模型体系构建
数据分层场景
分层混合效应模型
适用多工艺
高维稀疏小样本场景
差分拟合算法
适用刻蚀工艺
强非线性场景
高斯过程回归
适用薄膜工艺
算法 适应性设计
差分
拟合
特征
构造
邻域
分析
梯度
学习
模式
挖掘
差分拟合 算法 -跨刻蚀工艺的通用建模能力验证 刻蚀工艺的通用建模能力验证 刻蚀工艺的通用建模能力验证 刻蚀工艺的通用建模能力验证 刻蚀工艺的通用建模能力验证 刻蚀工艺的通用建模能力验证 刻蚀工艺的通用建模能力验证 刻蚀工艺的通用建模能力验证 刻蚀工艺的通用建模能力验证
Gaussian aussian aussian Process rocess rocess Regression egression egression egression
PECVD NDCPECVD NDCPECVD NDC PECVD NDC PECVD NDC
ALD ALD TiNTiN
PVD TTN PVD TTN PVD TTN PVD TTN PVD TTN PVD TTN
高斯过程 回归 -跨薄膜工艺的通用建模能力验证 薄膜工艺的通用建模能力验证 薄膜工艺的通用建模能力验证 薄膜工艺的通用建模能力验证 薄膜工艺的通用建模能力验证 薄膜工艺的通用建模能力验证 薄膜工艺的通用建模能力验证 薄膜工艺的通用建模能力验证 薄膜工艺的通用建模能力验证
算法通用性 验证
M1 -Al
ATAATA
DualDual -finfinfin
TRN TRN TRN TRN
PVD PVD PVD PVD TiNTiNTiN
ALDALD TMA TMA TMA TMA
PVD Cu Cu
N4 -CH
D4CD4CD4C-AT104AT104AT104AT104AT104
M3 -BWAIOBWAIOBWAIOBWAIOBWAIO
12KA12KA12KA12KA
W
TrenchTrench Trench
STISTISTI
GAA GAA
8
让价值 创造者实现更高让价值
创造者实现更高三、团队工作创新 – 建模流程标准化
模型可复用、扩展制
模块设计 流程建模 灵活 复用
模型 可解释 组件 高扩展
组件化建模
平台化架构
9
让价值 创造者实现更高让价值
创造者实现更高四、 对华创软件发展 的影响和贡献
NAURA Confidential
智能 推荐优化
多目标平衡优化 机理边界约束 可落地配方生成
参数 敏感度分析
多维分析建模 关键因子识别 调参方向明确
Chamber match Chamber match Chamber matchChamber match
偏差识别补偿 指标快速对齐 新腔快速迁移
• 模拟调参，避免盲试
指标预测
• 关键因子识别， 敏感度分析
工艺调参
• 参数空间高效寻优
智能推荐
• 腔室偏差学习，快速迁移
腔室对齐
• 经验标准化
，规律数字经验沉淀
• 验证数据回流，越用准
闭环学习
RefRefRef: AppliedPROAppliedPRO AppliedPROAppliedPROAppliedPRO AppliedPROAppliedPRO: process recipe optimizer for R&D acceleration : process recipe optimizer for R&D acceleration : process recipe optimizer for R&D acceleration : process recipe optimizer for R&D acceleration : process recipe optimizer for R&D acceleration : process recipe optimizer for R&D acceleration : process recipe optimizer for R&D acceleration : process recipe optimizer for R&D acceleration : process recipe optimizer for R&D acceleration : process recipe optimizer for R&D acceleration : process recipe optimizer for R&D acceleration : process recipe optimizer for R&D acceleration : process recipe optimizer for R&D acceleration : process recipe optimizer for R&D acceleration : process recipe optimizer for R&D acceleration : process recipe optimizer for R&D acceleration : process recipe optimizer for R&D acceleration : process recipe optimizer for R&D acceleration : process recipe optimizer for R&D acceleration : process recipe optimizer for R&D acceleration : process recipe optimizer for R&D acceleration : process recipe optimizer for R&D acceleration : process recipe optimizer for R&D acceleration : process recipe optimizer for R&D acceleration : process recipe optimizer for R&D acceleration : process recipe optimizer for R&D acceleration : process recipe optimizer for R&D acceleration : process recipe optimizer for R&D acceleration : process recipe optimizer for R&D acceleration : process recipe optimizer for R&D acceleration : process recipe optimizer for R&D acceleration : process recipe optimizer for R&D acceleration and beyond and beyond and beyond and beyond and beyond and beyond and beyond and beyond and beyond and beyond and beyond and beyond and beyond and beyond 论文 仅说明训练集 R² 为97% ，spec spec优化幅度 73 %
国际厂商 A AppliedProAppliedPro AppliedPro AppliedPro对比
从经验 驱动到 数据驱动 ，实现工艺经验 ，实现工艺经验 数字化 数字化 ，工艺指标 ，工艺指标 可预测 ，工艺模型 ，工艺模型 可复制迁移 可复制迁移 可复制迁移 ，沉淀 ，沉淀 工艺核心资产 工艺核心资产 工艺核心资产
10
让价值 创造者实现更高让价值
创造者实现更高五、 跨体系工作协同情况
NAURA Confidential
ConfidentialNAURA Confidential NAURA ConfidentialNAURA ConfidentialNAURA ConfidentialNAURA ConfidentialNAURA Confidential NAURA ConfidentialNAURA Confidential
二刻部署：
RecipeRecipeRecipeRecipe 数量
763 个
工艺覆盖种类
15 类
项目组已验证
2类
BU 工艺已验证
1类
BU 自主使用平台
12 类
数据收集及验证情况：
薄膜部署：
数据收集及验证情况 ：
RecipeRecipeRecipeRecipe 数量
322 个
工艺覆盖种类
6类
项目组已验证
2类
BU 工艺已验证
1类
BU 自主使用平台
3类
N1 二刻工艺 经理 反馈
工艺 配方智能优化平台 配方智能优化平台 配方智能优化平台 V1.0V1.0V1.0 在二刻上线后，平台功能的跨工艺应用力得到实际验证。支持蚀数据标 准化管理够将关键指在二刻上线后，平台功能的跨工艺应用力得到实际验证。支持蚀数据标 准化管理够将关键指在二刻上线后，平台功能的跨工艺应用力得到实际验证。支持蚀数据标 准化管理够将关键指在二刻上线后，平台功能的跨工艺应用力得到实际验证。支持蚀数据标 准化管理够将关键指在二刻上线后，平台功能的跨工艺应用力得到实际验证。支持蚀数据标 准化管理够将关键指在二刻上线后，平台功能的跨工艺应用力得到实际验证。支持蚀数据标 准化管理够将关键指在二刻上线后，平台功能的跨工艺应用力得到实际验证。支持蚀数据标 准化管理够将关键指在二刻上线后，平台功能的跨工艺应用力得到实际验证。支持蚀数据标 准化管理够将关键指在二刻上线后，平台功能的跨工艺应用力得到实际验证。支持蚀数据标 准化管理够将关键指在二刻上线后，平台功能的跨工艺应用力得到实际验证。支持蚀数据标 准化管理够将关键指在二刻上线后，平台功能的跨工艺应用力得到实际验证。支持蚀数据标 准化管理够将关键指在二刻上线后，平台功能的跨工艺应用力得到实际验证。支持蚀数据标 准化管理够将关键指在二刻上线后，平台功能的跨工艺应用力得到实际验证。支持蚀数据标 准化管理够将关键指在二刻上线后，平台功能的跨工艺应用力得到实际验证。支持蚀数据标 准化管理够将关键指在二刻上线后，平台功能的跨工艺应用力得到实际验证。支持蚀数据标 准化管理够将关键指在二刻上线后，平台功能的跨工艺应用力得到实际验证。支持蚀数据标 准化管理够将关键指在二刻上线后，平台功能的跨工艺应用力得到实际验证。支持蚀数据标 准化管理够将关键指在二刻上线后，平台功能的跨工艺应用力得到实际验证。支持蚀数据标 准化管理够将关键指在二刻上线后，平台功能的跨工艺应用力得到实际验证。支持蚀数据标 准化管理够将关键指在二刻上线后，平台功能的跨工艺应用力得到实际验证。支持蚀数据标 准化管理够将关键指在二刻上线后，平台功能的跨工艺应用力得到实际验证。支持蚀数据标 准化管理够将关键指在二刻上线后，平台功能的跨工艺应用力得到实际验证。支持蚀数据标 准化管理够将关键指在二刻上线后，平台功能的跨工艺应用力得到实际验证。支持蚀数据标 准化管理够将关键指在二刻上线后，平台功能的跨工艺应用力得到实际验证。支持蚀数据标 准化管理够将关键指在二刻上线后，平台功能的跨工艺应用力得到实际验证。支持蚀数据标 准化管理够将关键指在二刻上线后，平台功能的跨工艺应用力得到实际验证。支持蚀数据标 准化管理够将关键指与工艺参数的响应关系可视化， 帮助程师迅速锁定键影因子据分析效率显著提升。自研差拟合算法针对刻蚀高维小样本痛点与工艺参数的响应关系可视化， 帮助程师迅速锁定键影因子据分析效率显著提升。自研差拟合算法针对刻蚀高维小样本痛点与工艺参数的响应关系可视化， 帮助程师迅速锁定键影因子据分析效率显著提升。自研差拟合算法针对刻蚀高维小样本痛点与工艺参数的响应关系可视化， 帮助程师迅速锁定键影因子据分析效率显著提升。自研差拟合算法针对刻蚀高维小样本痛点与工艺参数的响应关系可视化， 帮助程师迅速锁定键影因子据分析效率显著提升。自研差拟合算法针对刻蚀高维小样本痛点与工艺参数的响应关系可视化， 帮助程师迅速锁定键影因子据分析效率显著提升。自研差拟合算法针对刻蚀高维小样本痛点与工艺参数的响应关系可视化， 帮助程师迅速锁定键影因子据分析效率显著提升。自研差拟合算法针对刻蚀高维小样本痛点与工艺参数的响应关系可视化， 帮助程师迅速锁定键影因子据分析效率显著提升。自研差拟合算法针对刻蚀高维小样本痛点与工艺参数的响应关系可视化， 帮助程师迅速锁定键影因子据分析效率显著提升。自研差拟合算法针对刻蚀高维小样本痛点与工艺参数的响应关系可视化， 帮助程师迅速锁定键影因子据分析效率显著提升。自研差拟合算法针对刻蚀高维小样本痛点与工艺参数的响应关系可视化， 帮助程师迅速锁定键影因子据分析效率显著提升。自研差拟合算法针对刻蚀高维小样本痛点与工艺参数的响应关系可视化， 帮助程师迅速锁定键影因子据分析效率显著提升。自研差拟合算法针对刻蚀高维小样本痛点与工艺参数的响应关系可视化， 帮助程师迅速锁定键影因子据分析效率显著提升。自研差拟合算法针对刻蚀高维小样本痛点与工艺参数的响应关系可视化， 帮助程师迅速锁定键影因子据分析效率显著提升。自研差拟合算法针对刻蚀高维小样本痛点与工艺参数的响应关系可视化， 帮助程师迅速锁定键影因子据分析效率显著提升。自研差拟合算法针对刻蚀高维小样本痛点与工艺参数的响应关系可视化， 帮助程师迅速锁定键影因子据分析效率显著提升。自研差拟合算法针对刻蚀高维小样本痛点与工艺参数的响应关系可视化， 帮助程师迅速锁定键影因子据分析效率显著提升。自研差拟合算法针对刻蚀高维小样本痛点与工艺参数的响应关系可视化， 帮助程师迅速锁定键影因子据分析效率显著提升。自研差拟合算法针对刻蚀高维小样本痛点与工艺参数的响应关系可视化， 帮助程师迅速锁定键影因子据分析效率显著提升。自研差拟合算法针对刻蚀高维小样本痛点与工艺参数的响应关系可视化， 帮助程师迅速锁定键影因子据分析效率显著提升。自研差拟合算法针对刻蚀高维小样本痛点与工艺参数的响应关系可视化， 帮助程师迅速锁定键影因子据分析效率显著提升。自研差拟合算法针对刻蚀高维小样本痛点与工艺参数的响应关系可视化， 帮助程师迅速锁定键影因子据分析效率显著提升。自研差拟合算法针对刻蚀高维小样本痛点与工艺参数的响应关系可视化， 帮助程师迅速锁定键影因子据分析效率显著提升。自研差拟合算法针对刻蚀高维小样本痛点与工艺参数的响应关系可视化， 帮助程师迅速锁定键影因子据分析效率显著提升。自研差拟合算法针对刻蚀高维小样本痛点与工艺参数的响应关系可视化， 帮助程师迅速锁定键影因子据分析效率显著提升。自研差拟合算法针对刻蚀高维小样本痛点与工艺参数的响应关系可视化， 帮助程师迅速锁定键影因子据分析效率显著提升。自研差拟合算法针对刻蚀高维小样本痛点与工艺参数的响应关系可视化， 帮助程师迅速锁定键影因子据分析效率显著提升。自研差拟合算法针对刻蚀高维小样本痛点与工艺参数的响应关系可视化， 帮助程师迅速锁定键影因子据分析效率显著提升。自研差拟合算法针对刻蚀高维小样本痛点与工艺参数的响应关系可视化， 帮助程师迅速锁定键影因子据分析效率显著提升。自研差拟合算法针对刻蚀高维小样本痛点有效根据工艺目标推荐更优参数组合 有效根据工艺目标推荐更优参数组合 有效根据工艺目标推荐更优参数组合 有效根据工艺目标推荐更优参数组合 有效根据工艺目标推荐更优参数组合 。
在NCHMNCHM 工艺中 工艺中 ，MCDMCD 目标为 目标为 50nm50nm50nm ，推荐参数跑片结果由 ，推荐参数跑片结果由 ，推荐参数跑片结果由 ，推荐参数跑片结果由 ，推荐参数跑片结果由 51.751.751.7 优化至 优化至 49.249.249.2 ；BCD 目标为 目标为 50nm50nm50nm ，由 52.852.852.852.8优化至 优化至 50 ；LER LER目标为 目标为 1.21.2 ，由 1.471.47 1.47优化至 优化至 1.211.21 ；SI -arc remainarc remainarc remain arc remainarc remain arc remainarc remainarc remain目标为 20 ，由 18.618.618.6 优化至 优化至 19.519.519.5 。在 8个目标 个目标 spec 中，推荐配方跑片 中，推荐配方跑片 中，推荐配方跑片 中，推荐配方跑片 中，推荐配方跑片 从base recipe base recipe base recipebase recipe base recipe 的2个specspecspec 命中提升至 命中提升至 命中提升至 6个命中 ，整体 形貌得 形貌得 到优化 。在 28HK28HK28HK28HK、AR 、TrenchTrenchTrench 等典型工艺中，多目标的关键参数 等典型工艺中，多目标的关键参数 等典型工艺中，多目标的关键参数 等典型工艺中，多目标的关键参数 等典型工艺中，多目标的关键参数 等典型工艺中，多目标的关键参数 等典型工艺中，多目标的关键参数 等典型工艺中，多目标的关键参数 trendtrendtrend 分析符合工艺认知，参数影响突破传统单的局限助力程 分析符合工艺认知，参数影响突破传统单的局限助力程 分析符合工艺认知，参数影响突破传统单的局限助力程 分析符合工艺认知，参数影响突破传统单的局限助力程 分析符合工艺认知，参数影响突破传统单的局限助力程 分析符合工艺认知，参数影响突破传统单的局限助力程 分析符合工艺认知，参数影响突破传统单的局限助力程 分析符合工艺认知，参数影响突破传统单的局限助力程 分析符合工艺认知，参数影响突破传统单的局限助力程 分析符合工艺认知，参数影响突破传统单的局限助力程 分析符合工艺认知，参数影响突破传统单的局限助力程 分析符合工艺认知，参数影响突破传统单的局限助力程 分析符合工艺认知，参数影响突破传统单的局限助力程 分析符合工艺认知，参数影响突破传统单的局限助力程 分析符合工艺认知，参数影响突破传统单的局限助力程 分析符合工艺认知，参数影响突破传统单的局限助力程 师提升工艺规律挖掘与推优验证效率。 师提升工艺规律挖掘与推优验证效率。 师提升工艺规律挖掘与推优验证效率。 师提升工艺规律挖掘与推优验证效率。 师提升工艺规律挖掘与推优验证效率。 师提升工艺规律挖掘与推优验证效率。 师提升工艺规律挖掘与推优验证效率。 师提升工艺规律挖掘与推优验证效率。 师提升工艺规律挖掘与推优验证效率。 师提升工艺规律挖掘与推优验证效率。
CEO -officeofficeofficeofficeofficeoffice工艺专家反馈 工艺专家反馈
• Platform Application Value and Overview: Platform Application Value and Overview: Platform Application Value and Overview:Platform Application Value and Overview: Platform Application Value and Overview:Platform Application Value and Overview: Platform Application Value and Overview: Platform Application Value and Overview: Platform Application Value and Overview:Platform Application Value and Overview: Platform Application Value and Overview:Platform Application Value and Overview:Platform Application Value and Overview: Platform Application Value and Overview: Platform Application Value and Overview:
The The The SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching SW is developed to serve practical needs in principle and iterated many cycles with the feedback validation from etching pro cess cess cess cess team. The platform realizes a complete close team. The platform realizes a complete close team. The platform realizes a complete closeteam. The platform realizes a complete closeteam. The platform realizes a complete close team. The platform realizes a complete close team. The platform realizes a complete closeteam. The platform realizes a complete close team. The platform realizes a complete closeteam. The platform realizes a complete close team. The platform realizes a complete closeteam. The platform realizes a complete close team. The platform realizes a complete closeteam. The platform realizes a complete close team. The platform realizes a complete close team. The platform realizes a complete close team. The platform realizes a complete close-loop from data upload to parameter optimization. In particular, the sen loop from data upload to parameter optimization. In particular, the sen loop from data upload to parameter optimization. In particular, the sen loop from data upload to parameter optimization. In particular, the sen loop from data upload to parameter optimization. In particular, the senloop from data upload to parameter optimization. In particular, the sen loop from data upload to parameter optimization. In particular, the sen loop from data upload to parameter optimization. In particular, the sen loop from data upload to parameter optimization. In particular, the senloop from data upload to parameter optimization. In particular, the sen loop from data upload to parameter optimization. In particular, the sen loop from data upload to parameter optimization. In particular, the sen loop from data upload to parameter optimization. In particular, the sen loop from data upload to parameter optimization. In particular, the sen loop from data upload to parameter optimization. In particular, the sen loop from data upload to parameter optimization. In particular, the senloop from data upload to parameter optimization. In particular, the senloop from data upload to parameter optimization. In particular, the sen loop from data upload to parameter optimization. In particular, the sen loop from data upload to parameter optimization. In particular, the sen loop from data upload to parameter optimization. In particular, the sen loop from data upload to parameter optimization. In particular, the sen loop from data upload to parameter optimization. In particular, the sen loop from data upload to parameter optimization. In particular, the sen loop from data upload to parameter optimization. In particular, the sen loop from data upload to parameter optimization. In particular, the senloop from data upload to parameter optimization. In particular, the sen sitivity sitivity sitivity analysis function analysis function analysis function analysis function analysis function breaks through the limitations of breaks through the limitations of breaks through the limitations of breaks through the limitations of breaks through the limitations of breaks through the limitations of breaks through the limitations of breaks through the limitations of breaks through the limitations of breaks through the limitations of breaks through the limitations of breaks through the limitations of breaks through the limitations of breaks through the limitations of breaks through the limitations of breaks through the limitations of breaks through the limitations of breaks through the limitations of breaks through the limitations of breaks through the limitations of breaks through the limitations of breaks through the limitations of breaks through the limitations of breaks through the limitations of breaks through the limitations of traditional singletraditional single traditional singletraditional singletraditional singletraditional singletraditional single traditional singletraditional single traditional single traditional single-parameter parameter parameter parameter parameter parameter analysis, significantly improving the efficiency of analysis, significantly improving the efficiency of analysis, significantly improving the efficiency of analysis, significantly improving the efficiency of analysis, significantly improving the efficiency of analysis, significantly improving the efficiency of analysis, significantly improving the efficiency of analysis, significantly improving the efficiency of analysis, significantly improving the efficiency of analysis, significantly improving the efficiency of analysis, significantly improving the efficiency of analysis, significantly improving the efficiency of analysis, significantly improving the efficiency of analysis, significantly improving the efficiency of analysis, significantly improving the efficiency of analysis, significantly improving the efficiency of analysis, significantly improving the efficiency of analysis, significantly improving the efficiency of analysis, significantly improving the efficiency of analysis, significantly improving the efficiency of analysis, significantly improving the efficiency of analysis, significantly improving the efficiency of analysis, significantly improving the efficiency of analysis, significantly improving the efficiency of analysis, significantly improving the efficiency of analysis, significantly improving the efficiency of analysis, significantly improving the efficiency of analysis, significantly improving the efficiency of analysis, significantly improving the efficiency of analysis, significantly improving the efficiency of analysis, significantly improving the efficiency of analysis, significantly improving the efficiency of analysis, significantly improving the efficiency of analysis, significantly improving the efficiency of analysis, significantly improving the efficiency of analysis, significantly improving the efficiency of analysis, significantly improving the efficiency of process rule mining, parameter optimization, and validationprocess rule mining, parameter optimization, and validation process rule mining, parameter optimization, and validation process rule mining, parameter optimization, and validation process rule mining, parameter optimization, and validationprocess rule mining, parameter optimization, and validationprocess rule mining, parameter optimization, and validationprocess rule mining, parameter optimization, and validationprocess rule mining, parameter optimization, and validation process rule mining, parameter optimization, and validation process rule mining, parameter optimization, and validationprocess rule mining, parameter optimization, and validationprocess rule mining, parameter optimization, and validation process rule mining, parameter optimization, and validationprocess rule mining, parameter optimization, and validationprocess rule mining, parameter optimization, and validationprocess rule mining, parameter optimization, and validation process rule mining, parameter optimization, and validationprocess rule mining, parameter optimization, and validationprocess rule mining, parameter optimization, and validationprocess rule mining, parameter optimization, and validationprocess rule mining, parameter optimization, and validationprocess rule mining, parameter optimization, and validationprocess rule mining, parameter optimization, and validationprocess rule mining, parameter optimization, and validation process rule mining, parameter optimization, and validationprocess rule mining, parameter optimization, and validationprocess rule mining, parameter optimization, and validation process rule mining, parameter optimization, and validationprocess rule mining, parameter optimization, and validation process rule mining, parameter optimization, and validation process rule mining, parameter optimization, and validationprocess rule mining, parameter optimization, and validation process rule mining, parameter optimization, and validationprocess rule mining, parameter optimization, and validationprocess rule mining, parameter optimization, and validation. The system provides a comprehensive and efficient tool for process The system provides a comprehensive and efficient tool for process The system provides a comprehensive and efficient tool for process The system provides a comprehensive and efficient tool for process The system provides a comprehensive and efficient tool for process The system provides a comprehensive and efficient tool for process The system provides a comprehensive and efficient tool for process The system provides a comprehensive and efficient tool for process The system provides a comprehensive and efficient tool for process The system provides a comprehensive and efficient tool for process The system provides a comprehensive and efficient tool for process The system provides a comprehensive and efficient tool for process The system provides a comprehensive and efficient tool for process The system provides a comprehensive and efficient tool for process The system provides a comprehensive and efficient tool for process The system provides a comprehensive and efficient tool for process The system provides a comprehensive and efficient tool for process The system provides a comprehensive and efficient tool for process The system provides a comprehensive and efficient tool for process The system provides a comprehensive and efficient tool for process The system provides a comprehensive and efficient tool for process The system provides a comprehensive and efficient tool for process The system provides a comprehensive and efficient tool for process The system provides a comprehensive and efficient tool for process The system provides a comprehensive and efficient tool for process The system provides a comprehensive and efficient tool for process The system provides a comprehensive and efficient tool for process The system provides a comprehensive and efficient tool for process The system provides a comprehensive and efficient tool for process The system provides a comprehensive and efficient tool for process The system provides a comprehensive and efficient tool for process engineers.
• Algorithm Implementation Innovation: Algorithm Implementation Innovation: Algorithm Implementation Innovation: Algorithm Implementation Innovation: Algorithm Implementation Innovation: Algorithm Implementation Innovation: Algorithm Implementation Innovation: Algorithm Implementation Innovation: Algorithm Implementation Innovation: Algorithm Implementation Innovation: Algorithm Implementation Innovation:
The algorithm team from the Digital Intelligence Department has independently developed a differential fitting algorithm. ThiThe algorithm team from the Digital Intelligence Department has independently developed a differential fitting algorithm. Thi The algorithm team from the Digital Intelligence Department has independently developed a differential fitting algorithm. Thi The algorithm team from the Digital Intelligence Department has independently developed a differential fitting algorithm. Thi The algorithm team from the Digital Intelligence Department has independently developed a differential fitting algorithm. Thi The algorithm team from the Digital Intelligence Department has independently developed a differential fitting algorithm. ThiThe algorithm team from the Digital Intelligence Department has independently developed a differential fitting algorithm. Thi The algorithm team from the Digital Intelligence Department has independently developed a differential fitting algorithm. Thi The algorithm team from the Digital Intelligence Department has independently developed a differential fitting algorithm. ThiThe algorithm team from the Digital Intelligence Department has independently developed a differential fitting algorithm. Thi The algorithm team from the Digital Intelligence Department has independently developed a differential fitting algorithm. Thi The algorithm team from the Digital Intelligence Department has independently developed a differential fitting algorithm. Thi The algorithm team from the Digital Intelligence Department has independently developed a differential fitting algorithm. ThiThe algorithm team from the Digital Intelligence Department has independently developed a differential fitting algorithm. Thi The algorithm team from the Digital Intelligence Department has independently developed a differential fitting algorithm. Thi The algorithm team from the Digital Intelligence Department has independently developed a differential fitting algorithm. Thi The algorithm team from the Digital Intelligence Department has independently developed a differential fitting algorithm. Thi The algorithm team from the Digital Intelligence Department has independently developed a differential fitting algorithm. Thi The algorithm team from the Digital Intelligence Department has independently developed a differential fitting algorithm. Thi The algorithm team from the Digital Intelligence Department has independently developed a differential fitting algorithm. Thi The algorithm team from the Digital Intelligence Department has independently developed a differential fitting algorithm. Thi The algorithm team from the Digital Intelligence Department has independently developed a differential fitting algorithm. Thi The algorithm team from the Digital Intelligence Department has independently developed a differential fitting algorithm. ThiThe algorithm team from the Digital Intelligence Department has independently developed a differential fitting algorithm. Thi The algorithm team from the Digital Intelligence Department has independently developed a differential fitting algorithm. Thi The algorithm team from the Digital Intelligence Department has independently developed a differential fitting algorithm. Thi The algorithm team from the Digital Intelligence Department has independently developed a differential fitting algorithm. Thi The algorithm team from the Digital Intelligence Department has independently developed a differential fitting algorithm. Thi The algorithm team from the Digital Intelligence Department has independently developed a differential fitting algorithm. Thi The algorithm team from the Digital Intelligence Department has independently developed a differential fitting algorithm. Thi The algorithm team from the Digital Intelligence Department has independently developed a differential fitting algorithm. Thi The algorithm team from the Digital Intelligence Department has independently developed a differential fitting algorithm. Thi The algorithm team from the Digital Intelligence Department has independently developed a differential fitting algorithm. Thi The algorithm team from the Digital Intelligence Department has independently developed a differential fitting algorithm. ThiThe algorithm team from the Digital Intelligence Department has independently developed a differential fitting algorithm. Thi The algorithm team from the Digital Intelligence Department has independently developed a differential fitting algorithm. Thi The algorithm team from the Digital Intelligence Department has independently developed a differential fitting algorithm. ThiThe algorithm team from the Digital Intelligence Department has independently developed a differential fitting algorithm. Thi s as as algorithm is lgorithm is lgorithm is lgorithm is wellwellwellwell-adapted to the scenario of high adapted to the scenario of highadapted to the scenario of highadapted to the scenario of high adapted to the scenario of highadapted to the scenario of highadapted to the scenario of highadapted to the scenario of highadapted to the scenario of highadapted to the scenario of highadapted to the scenario of highadapted to the scenario of high adapted to the scenario of highadapted to the scenario of highadapted to the scenario of highadapted to the scenario of highadapted to the scenario of highadapted to the scenario of highadapted to the scenario of highadapted to the scenario of high -dimensional parameters and tested in various etch process applications. dimensional parameters and tested in various etch process applications.dimensional parameters and tested in various etch process applications. dimensional parameters and tested in various etch process applications.dimensional parameters and tested in various etch process applications. dimensional parameters and tested in various etch process applications.dimensional parameters and tested in various etch process applications.dimensional parameters and tested in various etch process applications. dimensional parameters and tested in various etch process applications.dimensional parameters and tested in various etch process applications.dimensional parameters and tested in various etch process applications.dimensional parameters and tested in various etch process applications. dimensional parameters and tested in various etch process applications. dimensional parameters and tested in various etch process applications.dimensional parameters and tested in various etch process applications.dimensional parameters and tested in various etch process applications. dimensional parameters and tested in various etch process applications.dimensional parameters and tested in various etch process applications. dimensional parameters and tested in various etch process applications.dimensional parameters and tested in various etch process applications. dimensional parameters and tested in various etch process applications. dimensional parameters and tested in various etch process applications.dimensional parameters and tested in various etch process applications. dimensional parameters and tested in various etch process applications.dimensional parameters and tested in various etch process applications.dimensional parameters and tested in various etch process applications. dimensional parameters and tested in various etch process applications.dimensional parameters and tested in various etch process applications.dimensional parameters and tested in various etch process applications. dimensional parameters and tested in various etch process applications. dimensional parameters and tested in various etch process applications.dimensional parameters and tested in various etch process applications.dimensional parameters and tested in various etch process applications. dimensional parameters and tested in various etch process applications. It successfully overcomes It successfully overcomes It successfully overcomes It successfully overcomes It successfully overcomes It successfully overcomes It successfully overcomes It successfully overcomes It successfully overcomes It successfully overcomes It successfully overcomes It successfully overcomes It successfully overcomes the problem of traditional machine learning methods in complex scenarios and demonstrates strong technical innovation the problem of traditional machine learning methods in complex scenarios and demonstrates strong technical innovation the problem of traditional machine learning methods in complex scenarios and demonstrates strong technical innovation the problem of traditional machine learning methods in complex scenarios and demonstrates strong technical innovation the problem of traditional machine learning methods in complex scenarios and demonstrates strong technical innovation the problem of traditional machine learning methods in complex scenarios and demonstrates strong technical innovation the problem of traditional machine learning methods in complex scenarios and demonstrates strong technical innovation the problem of traditional machine learning methods in complex scenarios and demonstrates strong technical innovation andthe problem of traditional machine learning methods in the complex scenarios and demonstrates strong technical innovation the problem of traditional machine learning methods in complex scenarios and demonstrates strong technical innovation andthe problem of traditional machine learning methods in the complex scenarios and demonstrates strong technical innovation the problem of traditional machine learning methods in complex scenarios and demonstrates strong technical innovation the problem of traditional machine learning methods in complex scenarios and demonstrates strong technical innovation andthe problem of traditional machine learning methods in the complex scenarios and demonstrates strong technical innovation the problem of traditional machine learning methods in complex scenarios and demonstrates strong technical innovation the problem of traditional machine learning methods in complex scenarios and demonstrates strong technical innovation andthe problem of traditional machine learning methods in the complex scenarios and demonstrates strong technical innovation the problem of traditional machine learning methods in complex scenarios and demonstrates strong technical innovation andthe problem of traditional machine learning methods in the complex scenarios and demonstrates strong technical innovation the problem of traditional machine learning methods in complex scenarios and demonstrates strong technical innovation the problem of traditional machine learning methods in complex scenarios and demonstrates strong technical innovation the problem of traditional machine learning methods in complex scenarios and demonstrates strong technical innovation the problem of traditional machine learning methods in complex scenarios and demonstrates strong technical innovation andthe problem of traditional machine learning methods in the complex scenarios and demonstrates strong technical innovation the problem of traditional machine learning methods in complex scenarios and demonstrates strong technical innovation the problem of traditional machine learning methods in complex scenarios and demonstrates strong technical innovation andthe problem of traditional machine learning methods in the complex scenarios and demonstrates strong technical innovation the problem of traditional machine learning methods in complex scenarios and demonstrates strong technical innovation the problem of traditional machine learning methods in complex scenarios and demonstrates strong technical innovation the problem of traditional machine learning methods in complex scenarios and demonstrates strong technical innovation the problem of traditional machine learning methods in complex scenarios and demonstrates strong technical innovation andthe problem of traditional machine learning methods in the complex scenarios and demonstrates strong technical innovation andthe problem of traditional machine learning methods in the complex scenarios and demonstrates strong technical innovation the problem of traditional machine learning methods in complex scenarios and demonstrates strong technical innovation andthe problem of traditional machine learning methods in the complex scenarios and demonstrates strong technical innovation the problem of traditional machine learning methods in complex scenarios and demonstrates strong technical innovation the problem of traditional machine learning methods in complex scenarios and demonstrates strong technical innovation the problem of traditional machine learning methods in complex scenarios and demonstrates strong technical innovation the problem of traditional machine learning methods in complex scenarios and demonstrates strong technical innovation andthe problem of traditional machine learning methods in the complex scenarios and demonstrates strong technical innovation the problem of traditional machine learning methods in complex scenarios and demonstrates strong technical innovation the problem of traditional machine learning methods in complex scenarios and demonstrates strong technical innovation andthe problem of traditional machine learning methods in the complex scenarios and demonstrates strong technical innovation the problem of traditional machine learning methods in complex scenarios and demonstrates strong technical innovation andpr actical actical applicability. applicability.
• Team Team Team Performance & Project Significance: Performance & Project Significance: Performance & Project Significance:Performance & Project Significance: Performance & Project Significance: Performance & Project Significance: Performance & Project Significance: Performance & Project Significance: Performance & Project Significance: Performance & Project Significance: Performance & Project Significance: Performance & Project Significance: Performance & Project Significance: Performance & Project Significance:
The project team has deeply explored user needs and polished product functions with professional efficient collaboration The project team has deeply explored user needs and polished product functions with professional efficient collaboration The project team has deeply explored user needs and polished product functions with professional efficient collaboration The project team has deeply explored user needs and polished product functions with professional efficient collaboration The project team has deeply explored user needs and polished product functions with professional efficient collaboration The project team has deeply explored user needs and polished product functions with professional efficient collaboration The project team has deeply explored user needs and polished product functions with professional efficient collaboration The project team has deeply explored user needs and polished product functions with professional efficient collaboration The project team has deeply explored user needs and polished product functions with professional efficient collaboration The project team has deeply explored user needs and polished product functions with professional efficient collaboration The project team has deeply explored user needs and polished product functions with professional efficient collaboration The project team has deeply explored user needs and polished product functions with professional efficient collaboration The project team has deeply explored user needs and polished product functions with professional efficient collaboration The project team has deeply explored user needs and polished product functions with professional efficient collaboration The project team has deeply explored user needs and polished product functions with professional efficient collaboration The project team has deeply explored user needs and polished product functions with professional efficient collaboration The project team has deeply explored user needs and polished product functions with professional efficient collaboration The project team has deeply explored user needs and polished product functions with professional efficient collaboration The project team has deeply explored user needs and polished product functions with professional efficient collaboration The project team has deeply explored user needs and polished product functions with professional efficient collaboration The project team has deeply explored user needs and polished product functions with professional efficient collaboration The project team has deeply explored user needs and polished product functions with professional efficient collaboration The project team has deeply explored user needs and polished product functions with professional efficient collaboration The project team has deeply explored user needs and polished product functions with professional efficient collaboration The project team has deeply explored user needs and polished product functions with professional efficient collaboration The project team has deeply explored user needs and polished product functions with professional efficient collaboration The project team has deeply explored user needs and polished product functions with professional efficient collaboration The project team has deeply explored user needs and polished product functions with professional efficient collaboration The project team has deeply explored user needs and polished product functions with professional efficient collaboration The project team has deeply explored user needs and polished product functions with professional efficient collaboration The project team has deeply explored user needs and polished product functions with professional efficient collaboration The project team has deeply explored user needs and polished product functions with professional efficient collaboration The project team has deeply explored user needs and polished product functions with professional efficient collaboration The project team has deeply explored user needs and polished product functions with professional efficient collaboration The project team has deeply explored user needs and polished product functions with professional efficient collaboration The project team has deeply explored user needs and polished product functions with professional efficient collaboration The project team has deeply explored user needs and polished product functions with professional efficient collaboration The project team has deeply explored user needs and polished product functions with professional efficient collaboration The project team has deeply explored user needs and polished product functions with professional efficient collaboration The project team has deeply explored user needs and polished product functions with professional efficient collaboration The project team has deeply explored user needs and polished product functions with professional efficient collaboration The project team has deeply explored user needs and polished product functions with professional efficient collaboration The project team has deeply explored user needs and polished product functions with professional efficient collaboration The project team has deeply explored user needs and polished product functions with professional efficient collaboration The project team has deeply explored user needs and polished product functions with professional efficient collaboration thr oughout the oughout the process. By upgrading the etching process optimization from experience process. By upgrading the etching process optimization from experience process. By upgrading the etching process optimization from experience process. By upgrading the etching process optimization from experienceprocess. By upgrading the etching process optimization from experience process. By upgrading the etching process optimization from experienceprocess. By upgrading the etching process optimization from experience process. By upgrading the etching process optimization from experience process. By upgrading the etching process optimization from experienceprocess. By upgrading the etching process optimization from experience process. By upgrading the etching process optimization from experienceprocess. By upgrading the etching process optimization from experience process. By upgrading the etching process optimization from experience process. By upgrading the etching process optimization from experience process. By upgrading the etching process optimization from experience process. By upgrading the etching process optimization from experience process. By upgrading the etching process optimization from experience process. By upgrading the etching process optimization from experience process. By upgrading the etching process optimization from experienceprocess. By upgrading the etching process optimization from experienceprocess. By upgrading the etching process optimization from experience process. By upgrading the etching process optimization from experience process. By upgrading the etching process optimization from experience process. By upgrading the etching process optimization from experience process. By upgrading the etching process optimization from experience process. By upgrading the etching process optimization from experienceprocess. By upgrading the etching process optimization from experience -driven to data driven to data driven to data driven to datadriven to data -driven, the project not only provides driven, the project not only provides driven, the project not only provides driven, the project not only provides driven, the project not only provides driven, the project not only provides driven, the project not only provides driven, the project not only provides driven, the project not only provides driven, the project not only provides driven, the project not only provides driven, the project not only provides driven, the project not only provides an intelligent an intelligent an intelligent an intelligent an intelligent an intelligent an intelligent an intelligent tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields. tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields. tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields. tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields. tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields. tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields. tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields. tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields. tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields. tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields. tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields. tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields. tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields. tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields. tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields. tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields. tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields. tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields. tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields. tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields. tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields. tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields. tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields. tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.tool for the upgrading of etching processes but also exhibits extremely high promotion value in related fields.
N8 薄膜工艺 总监 反馈
平台针对薄膜沉积工艺实现标准化通用算法模型，在 平台针对薄膜沉积工艺实现标准化通用算法模型，在 平台针对薄膜沉积工艺实现标准化通用算法模型，在 平台针对薄膜沉积工艺实现标准化通用算法模型，在 平台针对薄膜沉积工艺实现标准化通用算法模型，在 平台针对薄膜沉积工艺实现标准化通用算法模型，在 平台针对薄膜沉积工艺实现标准化通用算法模型，在 平台针对薄膜沉积工艺实现标准化通用算法模型，在 平台针对薄膜沉积工艺实现标准化通用算法模型，在 ALD ALD ALD TiN 工艺场景下完成落地验证，推 工艺场景下完成落地验证，推 工艺场景下完成落地验证，推 工艺场景下完成落地验证，推 工艺场景下完成落地验证，推 优工艺 优工艺 参数组合跑片验证成功，结果符工艺要求。在 参数组合跑片验证成功，结果符工艺要求。在 参数组合跑片验证成功，结果符工艺要求。在 参数组合跑片验证成功，结果符工艺要求。在 参数组合跑片验证成功，结果符工艺要求。在 参数组合跑片验证成功，结果符工艺要求。在 参数组合跑片验证成功，结果符工艺要求。在 参数组合跑片验证成功，结果符工艺要求。在 参数组合跑片验证成功，结果符工艺要求。在 参数组合跑片验证成功，结果符工艺要求。在 参数组合跑片验证成功，结果符工艺要求。在 PVD TTN PVD TTN PVD TTNPVD TTNPVD TTN工艺中，平台实现 工艺中，平台实现 工艺中，平台实现 工艺中，平台实现 BIMBIM 、BOMBOMBOM对Rs 、RsURsU 的响应面分析，结果符合工艺预期能够有效帮助程师提升窗口率 的响应面分析，结果符合工艺预期能够有效帮助程师提升窗口率 的响应面分析，结果符合工艺预期能够有效帮助程师提升窗口率 的响应面分析，结果符合工艺预期能够有效帮助程师提升窗口率 的响应面分析，结果符合工艺预期能够有效帮助程师提升窗口率 的响应面分析，结果符合工艺预期能够有效帮助程师提升窗口率 的响应面分析，结果符合工艺预期能够有效帮助程师提升窗口率 的响应面分析，结果符合工艺预期能够有效帮助程师提升窗口率 的响应面分析，结果符合工艺预期能够有效帮助程师提升窗口率 的响应面分析，结果符合工艺预期能够有效帮助程师提升窗口率 的响应面分析，结果符合工艺预期能够有效帮助程师提升窗口率 的响应面分析，结果符合工艺预期能够有效帮助程师提升窗口率 的响应面分析，结果符合工艺预期能够有效帮助程师提升窗口率 的响应面分析，结果符合工艺预期能够有效帮助程师提升窗口率 的响应面分析，结果符合工艺预期能够有效帮助程师提升窗口率 的响应面分析，结果符合工艺预期能够有效帮助程师提升窗口率 的响应面分析，结果符合工艺预期能够有效帮助程师提升窗口率 的响应面分析，结果符合工艺预期能够有效帮助程师提升窗口率 的响应面分析，结果符合工艺预期能够有效帮助程师提升窗口率 。在 PECVD PECVD PECVD NDCNDCNDC工艺，完成针对机台漂移情况的补充实验设计 工艺，完成针对机台漂移情况的补充实验设计 工艺，完成针对机台漂移情况的补充实验设计 工艺，完成针对机台漂移情况的补充实验设计 工艺，完成针对机台漂移情况的补充实验设计 工艺，完成针对机台漂移情况的补充实验设计 工艺，完成针对机台漂移情况的补充实验设计 工艺，完成针对机台漂移情况的补充实验设计 工艺，完成针对机台漂移情况的补充实验设计 工艺，完成针对机台漂移情况的补充实验设计 工艺，完成针对机台漂移情况的补充实验设计 ，基于 ，基于 新机台跑片 新机台跑片 新机台跑片 新机台跑片 实验， 实验， 建立腔室迁移模型 建立腔室迁移模型 建立腔室迁移模型 建立腔室迁移模型 建立腔室迁移模型 ，模型 ，模型 推荐参数组合跑片符工艺目标要求。 推荐参数组合跑片符工艺目标要求。 推荐参数组合跑片符工艺目标要求。 推荐参数组合跑片符工艺目标要求。 推荐参数组合跑片符工艺目标要求。 推荐参数组合跑片符工艺目标要求。 推荐参数组合跑片符工艺目标要求。 推荐参数组合跑片符工艺目标要求。 推荐参数组合跑片符工艺目标要求。 通过在不同工艺场景中的验证， 通过在不同工艺场景中的验证， 通过在不同工艺场景中的验证， 通过在不同工艺场景中的验证， 平台具备贴合工艺研发的数据分析能力，模型推荐参组经验证均满足 平台具备贴合工艺研发的数据分析能力，模型推荐参组经验证均满足 平台具备贴合工艺研发的数据分析能力，模型推荐参组经验证均满足 平台具备贴合工艺研发的数据分析能力，模型推荐参组经验证均满足 平台具备贴合工艺研发的数据分析能力，模型推荐参组经验证均满足 平台具备贴合工艺研发的数据分析能力，模型推荐参组经验证均满足 平台具备贴合工艺研发的数据分析能力，模型推荐参组经验证均满足 平台具备贴合工艺研发的数据分析能力，模型推荐参组经验证均满足 平台具备贴合工艺研发的数据分析能力，模型推荐参组经验证均满足 平台具备贴合工艺研发的数据分析能力，模型推荐参组经验证均满足 平台具备贴合工艺研发的数据分析能力，模型推荐参组经验证均满足 平台具备贴合工艺研发的数据分析能力，模型推荐参组经验证均满足 平台具备贴合工艺研发的数据分析能力，模型推荐参组经验证均满足 平台具备贴合工艺研发的数据分析能力，模型推荐参组经验证均满足 平台具备贴合工艺研发的数据分析能力，模型推荐参组经验证均满足 平台具备贴合工艺研发的数据分析能力，模型推荐参组经验证均满足 目标要求 目标要求 ，可有效用于工艺推优 ，可有效用于工艺推优 ，可有效用于工艺推优 ，可有效用于工艺推优 ，可有效用于工艺推优 。
部署应用
11
让价值 创造者实现更高让价值
创造者实现更高六、价值观及关键事件
NAURA Confidential
以客户为中心
从工艺研发 需求出工艺研发
需求出贴合工艺分析场景
提升工艺研发效率 提升工艺研发效率
平台上线 挖掘 刻蚀 、薄膜工艺的 研发需求，搭建工艺 配方 智能优 化平台 V1.0V1.0V1.0V1.0并上线，实现数据 接入 →模型分析 →参数推优 →实验 验 证闭环。
工艺 支持 实现 算法 通用化设计 与算法组件可 复用，标准化算法可 跨工艺 支持 PVD PVD、CVDCVDCVD、ALDALD 、EtchEtchEtch 工艺的模型构建与推优。
业务 价值 拓展 工艺程师数据分析的空间维度， 提升数据分析效 率， 获得 BU 资深专家高度认可 。
持续创新
自研算法 解决刻蚀工艺数据高维度、 稀疏小样本的条件下， 传统机器学习算法失效的 难题，将验证工艺SPEC 命中 2个提升至 6个， 整体形貌得到 优化 。
漂移学习 针对不同腔室的工艺漂移情况， 针对不同腔室的工艺漂移情况， 构建工艺模型迁移策略， 在薄膜沉积工艺 实现基准模型迁移， 新腔 室跑片指标达率 100% 100% 。
业务 价值 深度挖掘工艺数据价值，拓宽调优边界标准化复 用工艺知识，提升研发效率。
突破传统算法局限 突破传统算法局限
适配工艺数据特点 适配工艺数据特点
落地工艺参数推优 落地工艺参数推优
北京方华创微电子装备有限公司
推动产业进步，创造无限可能 推动产业进步，创造无限可能 让价值创造者实现更高价值
