top of page
大會演講
-
Topic：Industrial Informatics and Intelligence in Shaping the Future of Digital Transformation
-
Speaker：Chair Prof. Fugee Tsung, HKUST
-
Abstract：
This seminar explores the pivotal role of industrial informatics and intelligence in driving digital transformation within the framework of Industry 4.0. It highlights how the integration of artificial intelligence, particularly through AI-generated content (AIGC) and large language models (LLMs), is reshaping interactions between technology and human expertise. Insights will be drawn from recent advancements at the Industrial Informatics and Intelligence Institute (Triple-I Institute) and the Quality and Data Analytics Lab.
-
Topic：Autonomous Robot Orchestration Solution for OHT with Machine Learning and Digital Twin
-
Speaker：Prof. Young Jae Jang, KAIST & CEO, DAIM Research
-
Abstract：
The Autonomous Robot Orchestration Solution (AROS) is transforming the management of robot fleets by enabling robots to collaboratively achieve shared objectives while adapting to their environment and operational state. This paper presents AROS and its application in controlling large fleets of Overhead Hoist Transport (OHT) vehicles within semiconductor fabrication facilities. AROS leverages advanced reinforcement learning algorithms and deep autoencoder models to optimize OHT operations. By integrating a Digital Twin (DT) system, which mirrors the real environment with real-time updates, AROS significantly enhances decision-making for OHT control. Our results demonstrate substantial improvements in OHT system performance, including reduced average delivery times and increased delivery capacity. Additionally, AROS incorporates robust anomaly detection and health monitoring solutions to address unexpected errors, effectively eliminating the need for manual intervention. The findings are validated through real-world implementations in large-scale semiconductor fabs.
-
Topic：A Framework to Implement Optimization Models in Semiconductor Supply Chains
-
Speaker：Prof. Sheng-I Chen, NYCU
-
Abstract：
This presentation focuses on production planning and scheduling problems with the development of optimization models to support decision-makers in semiconductor manufacturing.
We propose a highly modular framework that integrates an end-to-end process for implementing optimization software in organizations. This approach has significantly reduced the lead time for transforming business problems into optimization models. This potentially helps organizations adopt optimization easily and implement an optimization framework that can enhance production efficiency, improve inventory optimization, reduce waste, and minimize downtime.
Additionally, we explain mixed-integer linear programming formulations of classical problems and demonstrate how these models can be extended to help real-world analysis in the areas of job dispatching, capacity planning and production planning. Finally, various solution approaches are investigated to reduce run-time and obtain higher solution qualities which will improve prescriptive decision making that brings immediate business return.
-
Topic：Toward Digital and Sustainable Manufacturing with NVIDIA Omniverse
-
Speaker：Dr. Andrew Liu, NVIDIA
-
Abstract：
Industrial-scale digital twins powered by NVIDIA Omniverse and OpenUSD are revolutionizing manufacturing sustainability and scalability. By integrating data from various sources into a real-world context, they eliminate silos and provide a comprehensive view for cross-team collaboration. These digital twins allow engineers to simulate and optimize designs, layouts, energy usage, and processes before production begins, improving efficiency, reducing errors, and minimizing waste. For example, Siemens enhances this innovation by integrating NVIDIA Omniverse APIs into its Siemens Xcelerator platform, starting with Teamcenter X. This integration brings photorealistic visualization and physics-based digital twins to complex engineering workflows, further reducing waste and errors.
-
Topic：From Prescriptive Analytics to Optimization-Guided Learning
-
Speaker：Prof. Chia-Yen Lee, NTU
-
Abstract：
Typically, data analytics goes through phases such as descriptive analytics, diagnostic analytics, predictive analytics, and prescriptive analytics, all aimed at supporting a variety of decisions. This talk presents an optimization-guided learning framework that provides a feedback loop to enhance model training or analytics. Two methods were developed: genetic algorithm embedded with reinforcement learning (GAeRL) and reinforcement learning integrated with robust optimization (RLeRO). The proposed framework addresses the challenge of local optima in the model training process by incorporating optimization methods, thereby enhancing both the objective value and learning stability. Numerical studies on chemical production scheduling were conducted to validate the proposed framework.
-
Topic：Implementation Strategies for Decision-Making and Generative AI in Enterprises: Practical Approaches and Challenges
-
Speaker：Dr. Ben Lin, Profet AI
-
Abstract：
-
演講主題：未來光學世界 & 反敗為勝
-
演講貴賓：李天堯 玉晶光電 副總經理
-
演講摘要：
未來的光學世界
鏡頭開發：智能製造
反敗為勝
-
演講主題：AI與智慧製造: 智能工廠的發展趨勢
-
演講貴賓：蔡育典 副處長 (台積電十五B廠)
-
演講摘要：
AI(人工智慧)和智慧製造已成為半導體製造的核心技術，而工業工程(IE)專業在這個轉型中扮演了關鍵角色。台積電 Giga-Fab 結合大數據分析和人工智慧，使製造流程更加自動化及高效能。
工業工程的專業技能在智能工廠運作中發揮巨大影響，尤其是在流程優化、系統分析及資源配置等方面。
這次的演講將介紹在智慧製造中相關的AI應用，包括預測性維護、品質控制及生產排程等。透過AI的預測性維護，智能工廠可以有效預測設備故障風險，提前規劃維修，減少非計畫性停機；
在品質控制方面，AI使用影像識別技術快速檢測產品缺陷，提高生產品質和良率；在生產排程上，利用AI算法進行資源優化，實現高效排程。智能工廠中的AI應用為工業工程帶來了更多的挑戰與機會。隨著智慧製造發展，IE 結合 AI，將在未來提供更多可能性與競爭力。
指導單位： 主辦單位：
社團法人中國工業工程學會
國立臺北科技大學工業工程與管理系
協辦單位：
國立臺北科技大學管理學院
GUROBI Optimization
Japan Industrial Management Association
勞動及職業安全衛生研究所
台灣積體電路製造股份有限公司
中華民國人因工程學會
社團法人國立臺北科技大學工業工程與管理系所友會
杰倫智能科技
TAIA台灣人工智慧協會
中華萃思學會
bottom of page