Digitization and Data Analytics in Off-site Construction

场外施工的数字化和数据分析

• 批准号: RGPIN-2020-04126
• 负责人: Lei, Zhen
• 金额: $ 1.89万
• 依托单位: University of New Brunswick
• 依托单位国家: 加拿大
• 项目类别: Discovery Grants Program - Individual
• 财政年份: 2020
• 资助国家: 加拿大
• 起止时间: 2020-01-01 至 2021-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=716805
• 关键词: Digitization; Data; Analytics; Off; site

As an alternative approach to traditional stick-built construction methods, the off-site construction approach allows construction to be performed within a controlled environment, with components prefabricated and assembled into modules and units and then transported to construction sites for installation. However, as the traditional approach is replaced by off-site construction, construction data has become scattered throughout the supply chain, resulting in complex analytics. At this juncture, a new cognitive paradigm is required to understand the implementation of emerging data analytics techniques for offsite construction. In the manufacturing industry, rapid advancement of Industry 4.0 technologies, including cloud computing and the Internet of Things (IoT), has enabled in-depth data analysis through real-time data exchange. Machine learning and artificial intelligence (AI) techniques make it possible for autonomous decision-making such as proactive machine control and predictive maintenance. These advancements have disrupted manufacturing planning systems and human-machine interaction mechanisms. The digitization of the physical environment is often referred to as digital twinning, where seamless data exchange and decision-making protocols function simultaneously with industrial processes to create integrated cyber-physical systems. Recognizing the impact of digitization in the manufacturing industry, this research program will scientifically develop a framework to digitize the offsite construction supply chain and provide intelligent decision support tools for the construction industry. I plan to develop digital models (digital twins) for manufacturing facilities, transportation and delivery networks, and on-site construction assembly processes. Various decision-making protocols (namely predictive, simulation, and optimization models) will be tested and implemented in various applications. Through an immersive reality environment, I will explore how data-driven decisions can be translated to intuitive presentations to end-users in order to validate and improve decision support systems. The development of this research program will have a wide-reaching impact on construction project stakeholders by providing timely and accurate decision-making mechanisms. This will help boost construction productivity and, consequently, increase profitability of the Canadian construction industry. This research program will provide industry-engaging and cross-disciplinary training opportunities for Highly Qualified Personnel (HQP) at graduate and undergraduate levels. I will encourage and strive for Equity, Diversity and Inclusion (EDI) in my research program. The HQP will work closely with construction industry partners, collaborate with other research institutions/departments, and be exposed to interdisciplinary research topics and training.

作为传统棍式施工方法的替代方法，异地施工方法允许在受控环境中进行施工，预制组件并组装成模块和单元，然后运输到施工现场进行安装。然而，随着传统方法被异地施工取代，施工数据变得分散在整个供应链中，导致复杂的分析。在这一点上，需要一种新的认知范式来理解非现场建设的新兴数据分析技术的实施。在制造业，包括云计算和物联网(IoT)在内的Industry 4.0技术的快速发展，使人们能够通过实时数据交换进行深入的数据分析。机器学习和人工智能(AI)技术使主动机器控制和预测性维护等自主决策成为可能。这些进步已经扰乱了制造计划系统和人机交互机制。物理环境的数字化通常被称为数字结对，其中无缝的数据交换和决策协议与工业过程同时运行，以创建综合的网络-物理系统。 认识到数字化对制造业的影响，该研究计划将科学地制定一个框架，将非现场建筑供应链数字化，并为建筑业提供智能决策支持工具。我计划为制造设施、运输和交付网络以及现场施工组装过程开发数字模型(数字双胞胎)。各种决策协议(即预测、模拟和优化模型)将在各种应用中进行测试和实施。通过身临其境的现实环境，我将探索如何将数据驱动的决策转化为对最终用户的直观演示，以验证和改进决策支持系统。这项研究计划的开展将通过提供及时和准确的决策机制，对建设项目利益攸关方产生广泛影响。这将有助于提高建筑生产率，从而增加加拿大建筑业的盈利能力。 该研究计划将为研究生和本科生水平的高素质人才(HQP)提供行业参与和跨学科培训机会。我将在我的研究计划中鼓励和争取公平、多样性和包容性(EDI)。HQP将与建造业伙伴密切合作，与其他研究机构/部门合作，并参与跨学科研究课题和培训。