The full text of this item is not available at this time because the student has placed this item under an embargo for a period of time. The Libraries are not authorized to provide a copy of this work during the embargo period, even for Texas A&M users with NetID.
Data-Driven Approaches to Enhancing Highway Project Time Estimation
Abstract
State Highway Agencies (SHAs) collect a significant amount of digital data on highway projects during different project phases, including bid tabulations and Daily Work Reports
(DWRs). However, SHAs do not fully take advantage of the acquired digital data to obtain knowledge to apply to future projects. In the past decade, Artificial Intelligence (AI) techniques have rapidly been developed, offering promising data-driven solutions to recognizing complicated patterns from different types of data and detecting correlations between multiple variables. The goal of this dissertation is to create new knowledge, propose frameworks, and develop models to offer data-driven solutions to enhancing project time estimation in highway agencies.
In order to achieve the research goal, firstly, we explore the sequence logic of construction activities using data from DWRs adopting sequential Machine Learning (ML) techniques to identify project attributes affecting activity sequences and create an autonomous system for predicting activity sequences and overlapping in new projects. Secondly, since the work type of a project is an influential factor in project performance prediction in SHAs, the efficiency of current project work type classification is analyzed to identify drawbacks and offer a framework to classify projects into different work types effectively and identify key activities of each type. Toward this end, the data of bid tabulation and DWRs are utilized in unsupervised ML methods. Thirdly, an AI-based model is developed to take the cumulative activity progress in an arbitrary moment of highway projects to predict the remaining time of the project. In addition, the importance of different activities and project characteristics in determining the project time is identified.
The research contributes to the body of knowledge by finding, customizing, and fine-tuning advanced methods to leverage available digital data to obtain practical knowledge and create automated systems to support decision-making in terms of project schedule development, project categorizations, and project time estimation. SHAs in the US can use the findings of this research to improve the accuracy and reliability of their decisions in their highway projects.
Subject
Artificial IntelligenceTime Estimation
Sequence Analysis
Data Analytics
Construction Management
Classification
Regression
Deep Learning
Machine Learning
Citation
Alikhani, Hamed (2023). Data-Driven Approaches to Enhancing Highway Project Time Estimation. Doctoral dissertation, Texas A&M University. Available electronically from https : / /hdl .handle .net /1969 .1 /199100.