Project Details

Title:

A Machine Learning and Statistical Analysis Framework for Enhanced Engineer's Estimate Accuracy in Highway Infrastructure Projects, Phase I

Principal Investigators:

Ahmed Abdelaty

University:

University of Wyoming

Status:

Active

Type:

Research

Year:

2024

Grant #:

69A3552348308 (IIJA / BIL)

Project #:

CTIPS-015

RiP #:

01922769

RH Display ID:

161216

Keywords:

accuracy, bids, budgeting, construction projects, engineers, estimates, machine learning, statistical analysis

Abstract

State Transportation Agencies (STAs) rely on accurate engineer's estimates for budget allocation and contractor bid evaluation in highway projects. However, recent assessments reveal significant inaccuracies, with up to 25% deviations between engineers' estimates and awarded bids in the Wyoming Department of Transportation (WYDOT) in 2019. These deviations also resonate with similar findings published by other STAs. Challenges persist due to poor data quality and variations in estimating methods. This study aims to evaluate WYDOT's engineer's estimates' accuracy against historical bid data and assess consequences on project performance. Methodologically, a literature review and questionnaire survey will inform quantitative analysis of survey data and statistical analysis of bid tabulation data. By enhancing engineer's estimate accuracy, this research seeks to minimize budget deviations, improve project performance, and promote efficiency and transparency in transportation project planning and execution.

Project Word Files

project files

Project Description (Word, 163K)
UTC Project Information (Word, 86K)

Note to project PIs: The UTC document is limited to two pages. Also, it would be helpful if the Track Changes feature is used when editing either document above. Updated documents should be emailed to ndsu.ugpti@ndsu.edu.