Digitization of Scaffolding Management Processes in Construction Projects

在建筑工程中，脚手架经常被用来方便各种行业在架设或安装期间进行的工作。脚手架的预算可以占重工业建筑项目中预算的很大一部分（例如，项目直接工作预算总额的30 - 40%）。鉴于此，有效的脚手架活动规划对于项目的成功至关重要，忽视这些方面可能会导致预算超支和进度延迟。这需要准确的现场生产率指标，脚手架活动的基准数据，以及给定工作范围的脚手架工时估计。欣顿脚手架解决方案（安大略，加拿大）一直为重型和商业建筑项目提供脚手架服务，并努力寻找解决方案，以更好地评估，分析和预测现场性能。这导致了与新玩法大学合作的这个研究项目，利用机器学习（ML）算法来预测脚手架的生产力，以实现规划目的。具体来说，该研究项目将：（1）开发一个实时的物联网（IoT）脚手架生产率跟踪系统;（2）开发数据可视化和探索性数据分析（EDA）的仪表板系统，以带来业务洞察力;（3）开发贝叶斯回归模型，使用动态实时数据预测脚手架生产率。该项目将使合作伙伴公司在运营中变得更加灵活和更具成本效益。预计开发的系统将部署并应用于合作伙伴公司即将开展的项目。最后，该研究将通过引入建筑数据分析的新知识，使加拿大的建筑业受益。

In construction, scaffolding is often used to facilitate work performed by the various trades during erection or installation. The budget for scaffolding can account for a substantial amount of the budget in heavy industrial construction projects (e.g., 30 - 40% of the project's total direct work budget). Given this, efficient planning for scaffolding activities is essential for the project's success, and overlooking such aspects can result in budget overrun and schedule delay. This requires accurate field productivity metrics, benchmark data for the scaffolding activities, and estimates of scaffolding manhours for given scopes of work. Hinton Scaffolding Solutions (Ontario, Canada) has been providing scaffolding services for heavy and commercial construction projects, and strives to find solutions to better evaluate, analyze, and predict on-site performance. This leads to this research project in collaboration with the University of New Brunswick to utilize machine learning (ML) algorithms to predict scaffolding productivity for planning purposes. Specifically, this research project will: (1) develop a real-time, Internet-of-things (IoT) scaffolding productivity tracking system; (2) develop dashboard systems for data visualization and Exploratory Data Analysis (EDA) to bring business insights; and (3) develop Bayesian regression models to predict scaffolding productivity using dynamic, real-time data. This project will enable the partner company to become more agile and more cost-effective in their operations. The developed systems are expected to be deployed and applied to the partner company's upcoming projects. Lastly, the research will benefit Canada's construction industry by introducing new knowledge on construction data analytics.