CAREER: Knowledge Discovery in Databases and Data Mining as New Tools to Support Research and Educational Advances in Modern Construction Management

摘要CMS0094022“数据库中的知识发现和数据挖掘作为支持现代建筑管理研究和教育进步的新工具”，PI: Lucio Soibelman，伊利诺伊大学厄巴纳-香槟分校建筑行业在生成和收集数据的能力方面正在经历爆炸式的增长。科学数据收集的进步，几乎所有商业产品都采用条形码，新的传感器技术，无线计算和新的激光扫描技术，产生了大量的数据。这些进步加上数据存储技术的进步，例如更快、更高容量和更便宜的存储设备、更好的数据库管理系统和数据仓库技术，增加了计算机化建筑数据的可用性。然而，在大多数情况下，这些数据仅用于通信目的，并存储在文件或数据库中，而不进行分析。本项目旨在通过在数据库中应用数据挖掘和知识发现来研究这种不断增加的可用数据量。数据库中的知识发现和数据挖掘是结合机器学习、人工智能、模式识别、统计学、数据库和可视化等技术，从大型数据库中自动提取概念、概念相互关系和感兴趣的模式的技术。本CAREER项目的目标是：1)生成改进的方法，从大型建筑数据库中获取新知识，开发模型构建模板和向导，指导新手根据自己的数据创建模型；2)提高从业人员和学生对以往施工管理经验和知识的获取；3)运用主动学习技术，利用研究过程中产生的知识开发教育模拟游戏，提高各级学生的教育水平；4)通过数据挖掘、机器学习和人工智能工具的应用和开发，向土木与环境工程研究生讲授知识生成的过程。考虑到建筑行业在美国经济中的重要性，以及由于项目延迟、不切实际的预算、既不能满足质量要求又不能达到业绩预期的项目而在诉讼中浪费的大量资金，迫切需要改进管理工具。这项研究有望为改善项目规划和控制提供有价值的管理工具，如果应用于大型基础设施项目，可能会在全国范围内节省大量成本。这些研究成果可以推广到施工管理的各个子领域

ABSTRACT CMS0094022 "Knowledge Discovery in Databases and Data Mining as New tools to Support Research and Educational Advances in Modern Construction Management " PI: Lucio Soibelman, University of Illinois at Urbana-ChampaignThe construction industry is seeing an explosive growth in its capabilities to both generate and collect data. Advances in scientific data collection, the introduction of bar codes for almost all-commercial products, new sensor technologies, wireless computing, and new laser scanning technologies, have generated a flood of data. These advances coupled with advances in data storage technology, such as faster, higher capacity, and cheaper storage devices, better database management systems, and data warehousing technology, have increased the availability of computerized construction data. However, in most cases, these data are used only for communication purposes and stored in a file or a database without being analyzed. This project intends to study this increasing amount of available data by applying data mining and knowledge discovery in databases. Knowledge discovery in databases and data mining are technologies that combine techniques from machine learning, artificial intelligence, pattern recognition, statistics, databases and visualization to automatically extract concepts, concepts interrelationships, and patterns of interest from large databases. The objectives of this CAREER program are to: 1) Generate improved methods to obtain novel knowledge from large construction databases developing model-building templates and wizards to guide novice construction knowledge model builders through the process of creating models based on their own data; 2) Improve access to past construction management experience and knowledge by practitioners and students; 3) Use active learning techniques to improve education of students at all levels by developing an educational simulation game with the knowledge generated during this research; and 4) Teach civil and environmental engineering graduate students the process of knowledge generation through the application and development of data mining, machine learning and artificial intelligence tools.Given the importance of the construction industry in the U.S economy and the large amount of money wasted in litigation due to project delays, impractical budgets, and projects that neither satisfy quality requirements nor meet performance expectations, improved management tools are critically needed. This research promises to result in valuable management tools for improving project planning and control, which, if applied to large-scale infrastructure projects, may result in substantial cost savings nationwide. These research benefits can be extended to all sub-fields of construction management