Analyzing NPMRDS Using Advanced Machine Learning to Enhance Road Safety, Public Health and System Performance

Faezipour, Miad; Park, Sangho; Wu, Shuju

i

Analyzing NPMRDS Using Advanced Machine Learning to Enhance Road Safety, Public Health and System Performance

2025-09-12
By Faezipour, Miad ; Park, Sangho ; Wu, Shuju

File Language:

English

Details

Creators:

Faezipour, Miad ; Park, Sangho ; Wu, Shuju
Corporate Creators:

Central Connecticut State University ; Purdue University
Corporate Contributors:

Connecticut Department of Transportation ; United States. Department of Transportation. Federal Highway Administration
Subject/TRT Terms:

Data Collection Datasets Highway Safety Implementation Machine Learning National Performance Management Research Data Set (NPMRDS} Performance Public Health Traffic Speed
Publication/ Report Number:

CT-2337-F-25-4
Resource Type:

Tech Report
Geographical Coverage:

United States
Edition:

Final Report: 2004-2025
Corporate Publisher:

Connecticut. Department of Transportation
Abstract:

This project aims to leverage NPMRDS data and machine learning techniques to analyze the relationships between speed variations, crash occurrences, weather conditions, and road geometry. The goal is to identify patterns that can measure the performance of road segments and inform targeted safety interventions and infrastructure improvements. The project progressed in an organized sequence We began with systematic data collection and the selection of study segments, followed by rigorous data processing and feature engineering to support both speed prediction and roadway reliability assessment. Multiple modeling approaches are developed, validated, and refined to ensure robustness, while exploratory analyses are conducted to assess crash-related and health-related impacts, extending the scope of insights beyond prediction alone. The findings are carefully examined, leading to actionable recommendations for practice and policy. Implementation activities further demonstrate the practical applicability of the developed methods.
Format:

PDF
Funding:

SPR-2337
Collection(s):

US Transportation Collection
Main Document Checksum:

urn:sha-512:6f5a8840f0d6a9bb09ff9d6a14ebfd7a2f9c986f665909ae79e4027a33a47909e1b533aef4a36f4dd2ae8c5834aec9576597bf6ea11823a267a1270900e9fe1f
Download URL:

https://rosap.ntl.bts.gov/view/dot/88826/dot_88826_DS1.pdf
File Type:

[PDF - 768.89 KB ]

File Language:

English

ON THIS PAGE

Details

ROSA P serves as an archival repository of USDOT-published products including scientific findings, journal articles, guidelines, recommendations, or other information authored or co-authored by USDOT or funded partners. As a repository, ROSA P retains documents in their original published format to ensure public access to scientific information.