Level: Intermediate
TCM Section(s)
11.3. Information Management
7.2. Schedule Planning and Development
Venue: 2024 AACE International Conference & Expo
Abstract: This paper is an exploration and case study into the use of data analytics and machine learning as tools that can be combined with industry standards for schedule health analysis. Utilizing machine learning (ML) algorithms, schedule health data can predict future project schedule outcomes. The concepts presented are combined with current market tools for business intelligence reporting, scheduling, and model creation. The software used in this case study includes Microsoft Power BI, Oracle Primavera P6, and Microsoft AutoML to deliver high-quality schedule health data and forecasted performance on projects in a portfolio. The intent of using these techniques is to make data easily accessible and understandable to allow for enhanced decision-making capabilities throughout an organization.
This analysis supports AACE's proper schedule development adherence [1] and criteria for constructability reviews [2].The authors will show how machine learning is a natural next step in the data analytics process when building a schedule health dashboard. In this paper, the authors explain the typical machine learning process and compare it to Microsoft's AutoML machine learning tool, which is available as part of their Power BI analytics platform, where the advantages of AutoML are highlighted.
TCM Section(s)
11.3. Information Management
7.2. Schedule Planning and Development
Venue: 2024 AACE International Conference & Expo
Abstract: This paper is an exploration and case study into the use of data analytics and machine learning as tools that can be combined with industry standards for schedule health analysis. Utilizing machine learning (ML) algorithms, schedule health data can predict future project schedule outcomes. The concepts presented are combined with current market tools for business intelligence reporting, scheduling, and model creation. The software used in this case study includes Microsoft Power BI, Oracle Primavera P6, and Microsoft AutoML to deliver high-quality schedule health data and forecasted performance on projects in a portfolio. The intent of using these techniques is to make data easily accessible and understandable to allow for enhanced decision-making capabilities throughout an organization.
This analysis supports AACE's proper schedule development adherence [1] and criteria for constructability reviews [2].The authors will show how machine learning is a natural next step in the data analytics process when building a schedule health dashboard. In this paper, the authors explain the typical machine learning process and compare it to Microsoft's AutoML machine learning tool, which is available as part of their Power BI analytics platform, where the advantages of AutoML are highlighted.