Architectural Engineering and Construction, Durham School of

 

Durham School of Architectural Engineering and Construction: Faculty Publications

Accessibility Remediation

If you are unable to use this item in its current form due to accessibility barriers, you may request remediation through our remediation request form.

A Pattern-Recognition Method for Highway Construction Project Expenditure Cash Flows Using Clustering-Based K-Means Approach

Document Type

Article

Citation

Baek, M., Liang, Y., and Ashuri, B. “A Pattern-Recognition Method for Highway Construction Project Expenditure Cash Flows Using Clustering-Based K-Means Approach.” ASCE CRC, Arlington, VA, USA, March 9-12, 2022.

Abstract

An accurate cash flow model is essential because it helps state highway agencies forecast the funding obligations, reduce the financing burden, and streamline the delivery of their highway projects. Thus, the primary objective of this study is to discover the pattern of expenditure curves of highway projects and empirically explore the associations between affective factors and patterns of highway project expenditure curves. This paper analyzed the cash flows of 554 completed highway projects let in the State of Georgia. This study identifies five cash flow clusters using the unsupervised k-means clustering and analyzes the patterns of construction expenditure project cash flow with project-related attributes using multinomial logistic regression. Seven attributes, including owner estimate, number of pay items, and types of pay items, have statistically significant impacts on the shape of expenditure cash flow of highway projects. This study contributes to preparing more accurate estimates on expenditure cash flow for a project, making better letting/bidding decisions, and managing finance more effectively.

Share

COinS