Friday Transportation Seminars at Portland State University have been a tradition since 2000. We've opened up PSU Transportation Seminars to other days of the week, but the format is the same: Feel free to bring your lunch!  The seminar begins at 11:00 AM and goes until noon. If you can't join us in person, you can always watch online via Zoom: Register Here.

THE TOPIC 

Reproducibility is fundamental to the scientific process, allowing researchers and practitioners to build upon existing work, validate findings, and accelerate innovation. 

In transportation research, the inability to easily reproduce prior studies hinders progress, wasting valuable time and resources. Researchers often struggle to replicate results due to missing code, incomplete data, inadequate documentation, or incompatible computational environments. Despite the recognized importance of reproducibility, a comprehensive, quantitative assessment of its state within the transportation research domain is lacking. This deficiency limits the ability to track progress, identify best practices, and implement effective interventions to improve reproducibility.  

To address this critical gap, this project introduces and applies a scalable, automated methodology to systematically measure the prevalence of features associated with reproducibility in transportation research. We leverage large language models (LLMs) to extract these features from a large corpus of publications and their associated codebases. After careful validation of features extracted by LLMs, this innovative approach was used to analyze over 10,000 papers from leading transportation journals published between 2019 and 2024. 

This presentation will present a detailed quantitative analysis of the current state of reproducibility, offering insights into temporal trends, variations across journals and subfields, and the overall prevalence of open science practices within transportation research.

KEY LEARNING OUTCOMES

• A comprehensive, data-driven understanding of the current state of reproducibility and open science practices in transportation research.
• An innovative and scalable methodology that uses Large Language Models (LLMs) to automatically assess reproducibility features in scientific papers at a large scale.
• Specific insights into how reproducibility trends have evolved over the past five years and how practices vary between different journals and subfields.

SPEAKERS

Liming Wang, Portland State University

Dr. Liming Wang is a faculty member in the Toulan School of Urban Studies and Planning at Portland State University. While his primary research is in integrated land use and transportation modeling, he has recently caught the AI/LLM bug and is now exploring its application and impact on transportation research and practice.

PROFESSIONAL DEVELOPMENT

This 60-minute seminar is eligible for 1 hour of professional development credit for AICP (see our provider summary). We can provide an electronic attendance certificate for other types of certification maintenance.

ADD IT TO YOUR CALENDAR

Portland State University's Transportation Research and Education Center (TREC) is a multidisciplinary hub for all things transportation. We are home to the Initiative for Bicycle and Pedestrian Innovation (IBPI), the data programs PORTAL and BikePed Portal, the Better Block PSU program, and PSU's membership in PacTrans, the Pacific Northwest Transportation Consortium. Our continuing goal is to produce impactful research and tools for transportation decision makers, expand the diversity and capacity of the workforce, and engage students and professionals through education, seminars, and participation in research. To get updates about what's happening at TREC, sign up for our monthly newsletter or follow us on social media.