讲座一：Holistic Situational Awareness in Construction Site Safety Management
主讲嘉宾：Prof. Hubo Cai, Purdue University
讲座时间：2024年7月1日（周一）8：00
讲座地点：博学楼A100报告厅
腾讯会议：744-485-056
主办单位：工程经济与工程管理系、工程管理研究所
主讲嘉宾介绍：Dr. Hubo Cai is a Professor of Civil and Construction Engineering in the Lyles School of Civil Engineering and the Division of Construction Engineering and Management at Purdue University. His research focuses on infrastructure informatics at the intersection of data sensing, information modeling, computational applications, and risk management to improve the quality of infrastructure construction and enhance its lifecycle performance. He has published more than 100 technical reports and peer-reviewed journal and conference papers. He has over 20 years of experience in infrastructure engineering research and technology implementation and over three years of professional experience in infrastructure construction and asset management. Dr. Cai received his PhD and MS degrees in Civil Engineering from North Carolina State University, and a BS degree in Construction Management Engineering from Tongji University. He is a licensed Professional Engineer. Dr. Cai is an Associate Editor of the Journal of Computing in Civil Engineering.
讲座内容介绍：Construction industry has a poor safety record with a large number of fatalities and injuries. Construction typically accounts for 25% of all fatalities in the private sector. It is critical for construction workers to maintain situational awareness in the complex and dynamic job site. Towards this end, this seminar presents recent advances in sensing, deep learning algorithms, and visualization in assisting construction workers to develop and maintain their situational awareness in the challenging construction job sites. The talk summarizes a number of recently completed studies into three interrelated compartments: data-driven approach to holistic situational awareness for the prevention of struck-by accidents, real-time underground mapping and SmartDig to eliminate utility strikes, and ubiquitous sensing via EEG to identify job site hazards. The configuration of multimodal sensing systems, associated deep learning algorithms, and research findings will be described in detail.
讲座二：大语言模型在工程合同管理中的应用
主讲嘉宾：苏星, 浙江大学百人计划研究员
讲座时间：2024年7月1日（周一）10:30
讲座地点：博学楼A100报告厅
腾讯会议：744-485-056
主办单位：工程经济与工程管理系、工程管理研究所
主讲嘉宾介绍：苏星，浙江大学百人计划研究员，普渡大学博士，LEED AP， 浙江省土木建筑学会工程管理学术委员会副秘书长，美国Associated General Contractor (AGC)建筑信息科技委员会委员，BIM分委会委员。曾任华特迪士尼公司项目经理、南伊利诺伊大学助理教授。
讲座内容介绍：工程合同是项目利益相关方合作关系的基础，在保护双方权利与利益和项目的风险管理方面发挥着至关重要的作用。合同审查是控制风险发生和降低项目纠纷概率的有效手段，但目前的风险识别工作主要依靠专家，迫切需提升合同风险识别的自动化与智能化程度。大语言模型具备一定的推理能力，在工程合同管理尤其是风险识别方面具有应用潜力，但存在领域知识不足、易产生“幻觉”的问题。这里我们采用了知识增强技术，探索利用外部专家知识和事实知识来增强大语言模型风险识别表现的方法，并讨论相关技术的发展趋势和应用前景。
