<rss version="2.0" xmlns:atom="https://www.w3.org/2005/Atom">
  <channel>
    <title>Research in Progress (RIP)</title>
    <link>https://rip.trb.org/</link>
    <atom:link href="https://rip.trb.org/Record/RSS?s=PHNlYXJjaD48cGFyYW1zPjxwYXJhbSBuYW1lPSJkYXRlaW4iIHZhbHVlPSJhbGwiIC8+PHBhcmFtIG5hbWU9InN1YmplY3Rsb2dpYyIgdmFsdWU9Im9yIiAvPjxwYXJhbSBuYW1lPSJ0ZXJtc2xvZ2ljIiB2YWx1ZT0ib3IiIC8+PHBhcmFtIG5hbWU9ImxvY2F0aW9uIiB2YWx1ZT0iMTYiIC8+PC9wYXJhbXM+PGZpbHRlcnM+PGZpbHRlciBmaWVsZD0iaW5kZXh0ZXJtcyIgdmFsdWU9IiZxdW90O0RhdGEgYW5hbHlzaXMmcXVvdDsiIG9yaWdpbmFsX3ZhbHVlPSImcXVvdDtEYXRhIGFuYWx5c2lzJnF1b3Q7IiAvPjwvZmlsdGVycz48cmFuZ2VzIC8+PHNvcnRzPjxzb3J0IGZpZWxkPSJwdWJsaXNoZWQiIG9yZGVyPSJkZXNjIiAvPjwvc29ydHM+PHBlcnNpc3RzPjxwZXJzaXN0IG5hbWU9InJhbmdldHlwZSIgdmFsdWU9InB1Ymxpc2hlZGRhdGUiIC8+PC9wZXJzaXN0cz48L3NlYXJjaD4=" rel="self" type="application/rss+xml" />
    <description></description>
    <language>en-us</language>
    <copyright>Copyright © 2026. National Academy of Sciences. All rights reserved.</copyright>
    <docs>http://blogs.law.harvard.edu/tech/rss</docs>
    <managingEditor>tris-trb@nas.edu (Bill McLeod)</managingEditor>
    <webMaster>tris-trb@nas.edu (Bill McLeod)</webMaster>
    <image>
      <title>Research in Progress (RIP)</title>
      <url>https://rip.trb.org/Images/PageHeader-wTitle-RIP.jpg</url>
      <link>https://rip.trb.org/</link>
    </image>
    <item>
      <title>Data Driven Analysis and Modeling of Merging Behavior and Traffic Breakdown at Freeway Merging Bottlenecks</title>
      <link>https://rip.trb.org/View/2786711</link>
      <description><![CDATA[Freeway merging bottlenecks (e.g., lane-drop, entry ramps) are major sources of congestion, where traffic breakdown is often triggered by complex interactions among merging vehicles. While congestion has been widely studied, existing models for bottleneck congestion frequently rely on simplified behavioral assumptions and do not adequately capture the mechanisms that lead to congestion onset. This limits the reliability of simulation tools used for operational analysis and policy evaluation. This project proposes a data-driven, simulation-based framework to model merging behavior and quantify its role in congestion formation at merging bottlenecks. Using high-resolution drone-based trajectory data collected during congestion onset, the study will characterize key aspects of human merging behavior, including gap acceptance, lane-changing dynamics, and speed adaptation. Additional drone data collection will be conducted as needed to enrich the dataset. A probabilistic model will be developed to estimate the likelihood of congestion onset as a function of traffic conditions and merging behavior. This model will be integrated into a microscopic traffic simulation platform to reproduce realistic merging dynamics and congestion formation as an emergent phenomenon. The resulting framework will provide transportation agencies with a practical tool for analyzing congestion mechanisms and evaluating operational strategies. The project also establishes a foundation for future research on control strategies in mixed traffic systems, including automated vehicle applications.

]]></description>
      <pubDate>Mon, 05 Oct 2026 13:30:33 GMT</pubDate>
      <guid>https://rip.trb.org/View/2786711</guid>
    </item>
    <item>
      <title>Evaluation of Traffic Speed Deflection Devices (TSDD) for Collecting, Reporting, and Utilizing Pavement Network and Project Level Structural Data in Utah</title>
      <link>https://rip.trb.org/View/2784729</link>
      <description><![CDATA[Evaluate the potential of Traffic Speed Deflection Devices (TSDD) for collecting, reporting, and utilizing pavement network and project level structural data in Utah.]]></description>
      <pubDate>Wed, 30 Sep 2026 19:13:13 GMT</pubDate>
      <guid>https://rip.trb.org/View/2784729</guid>
    </item>
    <item>
      <title>Leveraging Location-Based Services (LBS) Data and Video Analytics for Scalable Intersection Safety Assessment</title>
      <link>https://rip.trb.org/View/2780208</link>
      <description><![CDATA[Intersections are high-risk locations where a significant share of severe crashes occur, particularly those involving pedestrians and other vulnerable road users. Traditional approaches to safety analysis rely on crash records, field observations, or fixed camera systems. While valuable, these methods are often costly, spatially limited, and retrospective, limiting their ability to support proactive, system-wide safety management. Transportation agencies increasingly require scalable, data-driven approaches to identify risky conditions and behaviors before crashes occur.

This project investigates whether anonymized Location-Based Services (LBS) data can be used to characterize and monitor safety-related conditions at urban intersections. LBS data provides large-scale coverage of mobility patterns but are inherently sparse and noisy. Rather than attempting event-level detection, this research focuses on whether LBS-derived movement indicators can capture safety-relevant conditions at the intersection level. The proposed approach leverages LBS data alongside complementary datasets, including video-detected safety events and crash records. The City of Seattle serves as the primary testbed due to the availability of processed traffic safety video data. These video data consist of event-triggered clips capturing safety-relevant interactions such as pedestrian signal violations, vehicle–pedestrian conflicts, and vehicles stopping within intersections. While limited in coverage, these data provide high-value examples of safety-critical behaviors.

A key component of the project is the systematic characterization of safety-related behaviors using video data. The video team will analyze a representative set of clips to refine and standardize behavior categories and develop a structured behavioral taxonomy. For each behavior type, key qualitative features will be identified, including temporal characteristics (e.g., timing and duration), movement characteristics (e.g., speed changes, trajectory deviations, stopping behavior), and interaction dynamics (e.g., number of agents and conflict interactions). These behavioral features will then be translated into measurable LBS-derived indicators, such as speed variability, trajectory density, spatial clustering, dwell time within intersections, and overlap of movement flows. This process will be conducted collaboratively to ensure that identified features are meaningful within the constraints of LBS data.

Due to the sampling and sparsity inherent in LBS data, individual events observed in video are unlikely to be directly captured in LBS trajectories. Therefore, the analysis does not attempt one-to-one event detection. Instead, the study evaluates whether LBS-derived indicators exhibit systematic variation across time periods and locations associated with differing levels of observed safety events. For example, locations or periods with elevated levels of pedestrian violations or vehicle–pedestrian conflicts may exhibit distinct LBS-derived indicators, such as higher trajectory density, clustering, or speed variability.

Because the usefulness of these indicators depends on data quality, the project will explicitly assess the effects of sampling density, spatial accuracy, temporal coverage, and observation frequency on indicator reliability. Intersections and analysis periods will be screened using minimum data quality thresholds to ensure sufficient trajectory coverage. The study will also identify conditions under which LBS-derived indicators become unreliable, such as low sample penetration, insufficient observations, or highly heterogeneous]]></description>
      <pubDate>Fri, 18 Sep 2026 10:38:20 GMT</pubDate>
      <guid>https://rip.trb.org/View/2780208</guid>
    </item>
    <item>
      <title>Integrating Transportation Data to Address Representation and Measurement Limitations in Travel Behavior and Demand Analysis</title>
      <link>https://rip.trb.org/View/2780201</link>
      <description><![CDATA[Expanding transportation data availability is creating new opportunities for a richer, potentially more representative, and broader data landscape to support the measurement and modeling of evolving travel behaviors and outcomes. This project addresses the need for clearer guidance on when, where, and how multiple data sources can be integrated to overcome common data limitations in travel behavior and demand research. The proposed work will develop an accessible taxonomy and application framework organized by stage of analysis – survey design, data preparation, analysis and modeling, validation, and calibration – and by the specific limitation being addressed: insufficient sample size, lack of population representativeness, unobserved key variables, and measurement error. By linking data limitations to appropriate integration strategies and transportation applications, this work will help move the field beyond isolated, ad hoc uses of data fusion toward a more systematic and transferable practice. The expected outcome is practical guidance that supports more credible, representative, and policy-relevant travel behavior and demand analysis in an increasingly complex data environment.]]></description>
      <pubDate>Fri, 18 Sep 2026 10:14:26 GMT</pubDate>
      <guid>https://rip.trb.org/View/2780201</guid>
    </item>
    <item>
      <title>Safe Driving Across Domains</title>
      <link>https://rip.trb.org/View/2778232</link>
      <description><![CDATA[This project tackles the critical research question: How can pre-trained autonomous driving models be seamlessly and dynamically adapted across diverse operational design domains (ODDs) to achieve safe and robust deployment? ODDs define the specific conditions under which an autonomous driving system can operate safely—spanning factors like road types, traffic density, weather patterns, and geographic features. For instance, a model trained to navigate Pittsburgh’s urban intersections must also adapt to Pittsburgh’s rural routes with sparse signage and rolling hills, or even underdeveloped, unpaved, mixed-traffic roads. What is the right way for a robot to, in real time, smoothly switch between different planning trajectories to ensure safety and consistency amid changing driving conditions?

By focusing on adapting pre-trained models rather than retraining them from scratch, this project offers a scalable approach to ensure safe deployment. This is particularly critical in data-limited settings—collecting training data for every possible scenario can cost millions and take years. For example, equipping autonomous systems to operate in rural areas avoids costly new data collection while preserving high safety standards. A single deployment failure in an untested environment could cost lives and erode public trust, emphasizing the importance of rigorous adaptation. With an estimated 1.35 million road traffic deaths annually worldwide in drastically different driving conditions, this project aims to ensure that autonomous driving technologies can be deployed safely. In other words, this research algorithmically and proactively catalyzes translation of new technologies to data-limited areas, instead of passively relying on commercial technology spillover – the latter would take years or even decades. 

The research team leverages diverse datasets to ensure robust testing. These include driving data from TIER IV in Japan, and over 10 million miles of publicly available driving data from Waymo, collected across 25 U.S. cities and capturing diverse urban environments. In Pittsburgh, the team has access to truck-mounted camera data that covers mixed urban and rural routes, while stationary cameras funded through a PITA partnership with RIDC capture real-time trajectory data in high-traffic industrial areas. Simulation-based fine-tuning complements these datasets, addressing data gaps in under-documented scenarios, such as rural roads with limited signage. This rich dataset ecosystem provides the foundation for testing and validating model adaptations in diverse operational design domains.

The project employs open source, cutting-edge algorithms, including inverse reinforcement learning (RL), guided diffusion, blueprints, and fine-tuning. Inverse RL provides an overall framework to complement rule-based modules to directly and flexibly extract algorithm parameters (sometimes implicitly) from expert (human or simulated) driving. Guided diffusion allows for precise and smooth generation and adjustments of such parameters in constrained domains, such as adapting autonomous driving algorithms from Pittsburgh's dense urban streets to its hilly rural roads. Blueprinting ensures scalable and structured design of template models that are transferable across diverse contexts, while fine-tuning efficiently adapts pre-trained models to novel environments, particularly in low-data settings. Open source models already implement these algorithms. The team adapts these powerful emerging methods to address challenges in rural areas and infrastructure-limited regions.]]></description>
      <pubDate>Thu, 17 Sep 2026 16:37:29 GMT</pubDate>
      <guid>https://rip.trb.org/View/2778232</guid>
    </item>
    <item>
      <title>AI-Enabled Aeromedical Safety Assessment Decision Support</title>
      <link>https://rip.trb.org/View/2778047</link>
      <description><![CDATA[This project will evaluate how artificial intelligence can support the Federal Aviation Administration's (FAA’s) Aeromedical Qualitative Hazard Assessment Framework while preserving human oversight, scientific rigor, transparency, traceability, and regulatory defensibility. The framework provides a structured method for assessing how medical conditions, treatments, and operational mitigations may affect pilot performance and aviation safety, but its application requires substantial expert effort.

The research will identify assessment activities suitable for artificial intelligence (AI) assistance, including evidence identification, impairment-pathway mapping, operational scenario development, bowtie modeling, barrier evaluation, confidence assessment, and report preparation. Candidate technologies—including large language models, knowledge-retrieval systems, and automated reasoning tools—will be compared with expert-led assessments using representative aeromedical policy cases. Evaluation will address accuracy, completeness, consistency, reproducibility, traceability, workload reduction, and timeliness.

The project will also examine potential biases, failure modes, evidence-quality concerns, and governance requirements. It will define the human-review responsibilities, technical architecture, data resources, validation methods, and controls needed for responsible operational use. Deliverables will include an assessment of AI applicability across the framework, a prototype AI-assisted workflow and decision-support architecture, case-study performance results, and an implementation roadmap. The findings will guide future FAA investments while ensuring that AI augments, rather than replaces, accountable expert judgment in aeromedical safety decisions.]]></description>
      <pubDate>Mon, 14 Sep 2026 20:55:05 GMT</pubDate>
      <guid>https://rip.trb.org/View/2778047</guid>
    </item>
    <item>
      <title>A Crowd-sourcing Approach to Developing Visual Datasets for Safety Assets</title>
      <link>https://rip.trb.org/View/2775968</link>
      <description><![CDATA[Recent advances in data management and data-driven analytics offer substantial potential to improve transportation safety asset management by automating labor-intensive condition assessment and data interpretation tasks. However, despite the rapid growth of imagery and sensor data within UDOT, most existing datasets are not directly usable for data-driven applications because they lack standardized, high-quality condition labels. Existing annotation efforts are largely manual, inconsistent, and difficult to sustain at scale, limiting the development and deployment of reliable computer vision and artificial intelligence (AI) tools.

This project addresses these limitations by developing a collaborative, web-based crowd-sourcing platform that enables UDOT personnel and approved stakeholders to efficiently annotate transportation safety asset imagery. The platform will incorporate standardized labeling workflows, quality assurance procedures, and automated metadata generation to create reusable, high-quality datasets that support both operational asset management and future AI applications. The resulting framework will establish a scalable process for transforming raw imagery into curated datasets suitable for inspection, maintenance prioritization, performance monitoring, and future research.]]></description>
      <pubDate>Thu, 10 Sep 2026 12:44:46 GMT</pubDate>
      <guid>https://rip.trb.org/View/2775968</guid>
    </item>
    <item>
      <title>Synthesis of Information Related to Transit Practices. Topic SG-23. Data-Driven Approaches for Optimizing Transit Parts Procurement, Lead Time Risk, and Critical Spare Parts Planning</title>
      <link>https://rip.trb.org/View/2772554</link>
      <description><![CDATA[Transit agencies depend on the timely availability of spare parts to maintain fleet reliability, minimize vehicle downtime, and sustain safe and efficient service. Agencies employ a variety of approaches to procure, stock, and manage spare parts inventories; however, practices vary considerably based on fleet composition, agency size, operating environment, warehouse capacity, staffing resources, and supplier relationships.

In recent years, transit agencies have experienced increasing challenges associated with extended supplier lead times, supply chain disruptions, parts obsolescence, and limited availability of critical vehicle components. At the same time, agencies are making greater use of maintenance management systems, enterprise resource planning (ERP) systems, inventory management software, vehicle diagnostic data, asset management programs, and other operational data to improve procurement and inventory decision-making.

Although many agencies have developed innovative practices to address these challenges, there has been limited documentation of current industry approaches for managing spare parts inventories, mitigating supply chain risks, identifying critical spare parts, and using data to support procurement decisions. A synthesis would provide transit agencies with a comprehensive understanding of current practices, emerging approaches, and lessons learned across the industry.

OBJECTIVE: The objective of this synthesis is to document current practices used by transit agencies to optimize spare parts procurement, inventory management, supplier lead time risk, and critical spare parts planning. The synthesis should identify how agencies use data, technology, and operational practices to improve parts availability, reduce vehicle downtime, and support reliable transit service.

]]></description>
      <pubDate>Tue, 01 Sep 2026 17:16:59 GMT</pubDate>
      <guid>https://rip.trb.org/View/2772554</guid>
    </item>
    <item>
      <title>Preparing the Airport Workforce for Emerging Technologies and Data-Informed Decision Making</title>
      <link>https://rip.trb.org/View/2772531</link>
      <description><![CDATA[Airports are operating in an increasingly complex environment driven by rapid technological change, expanding data availability, evolving safety expectations, and growing operational demands. At the same time, airports continue to face workforce challenges, including retirements, competition for skilled labor, recruitment and retention issues, and the loss of institutional knowledge. Although airports collect increasing amounts of workforce and operational data, these data often remain fragmented across separate systems, limiting their ability to align workforce capabilities and staffing decisions with operational needs and the adoption of emerging technologies.

The objective of this research is to develop practical guidance, tools, and implementation resources that help airports modernize workforce development while improving data-informed workforce decision-making. The research will produce an airport workforce framework, role-based training pathways, guidance for integrating workforce and operational data, and implementation resources that airports of varying size and complexity can use to strengthen workforce capabilities, support staffing and operational decisions, and complement existing workforce research while minimizing duplication.]]></description>
      <pubDate>Tue, 01 Sep 2026 10:02:54 GMT</pubDate>
      <guid>https://rip.trb.org/View/2772531</guid>
    </item>
    <item>
      <title>An Analysis of Crash Data and Safety Trends in ATV's and UTV's in Minnesota</title>
      <link>https://rip.trb.org/View/2762033</link>
      <description><![CDATA[All-terrain vehicles (ATVs) and utility task vehicles (UTVs) continue to remain under the radar as contributors to serious injuries and fatalities. In Minnesota, the number of deaths associated with ATVs averages about twenty annually. However, the number of serious injuries is even greater. Serious injury and fatal crashes have a significant impact not just on the individuals involved in the crashes, but their families and communities who bear the impacts of the aftereffects. Furthermore, these numbers are only those known to be reported in the motor vehicle crash data. There is the potential that the number of ATV- and UTV-involved crashes is greater when considering direct admissions to the hospital. Overall, the crash data trends are not fully understood. MnDOT is seeing an increasing trend in their occurrence, and they do not envision this trend de-escalating anytime soon. The topic of ATV/UTV has also been discussed with frequency in the Minnesota legislature. Therefore, this research seeks to better quantify the number and severity of crashes involving ATVs and UTVs as well as trends in this data. This in-depth analysis will methodically investigate the crash causals and circumstances that could then potentially inform education, policy, and perhaps engineering solutions to eliminate the serious injury and fatal crash occurrence.]]></description>
      <pubDate>Tue, 18 Aug 2026 16:42:44 GMT</pubDate>
      <guid>https://rip.trb.org/View/2762033</guid>
    </item>
    <item>
      <title>Advancing Geotechnical Site Investigation with Automated 3D SPT-Seismic Testing</title>
      <link>https://rip.trb.org/View/2762014</link>
      <description><![CDATA[Effective geotechnical site characterization is essential for the planning, design, and long-term performance of transportation infrastructure. Unanticipated subsurface conditions (such as buried voids, weak soils, or highly variable stratigraphy) can lead to costly construction delays, change orders, and, in severe cases, structural distress or failure. For State DOTs responsible for delivering safe and resilient infrastructure, obtaining reliable information on subsurface conditions and their spatial variability is therefore critical during the early stages of project development.

Current practice relies heavily on invasive point-based methods such as Standard Penetration Test (SPT), Cone Penetration Test (CPT), and rock coring. While these methods provide high-quality measurements, they sample only a very small volume of material within individual boreholes. As a result, large portions of the subsurface remain uncharacterized, and critical features (such as voids, soft zones, or irregular bedrock surfaces) remain undetected between boreholes. Surface-based geophysical methods can provide broader spatial coverage; however, their resolution typically decreases rapidly with depth due to signal attenuation, limiting their effectiveness for deeper subsurface imaging.

Recent advancements in seismic testing have led to the development of the 3D SPT-seismic method, an innovative approach that integrates seismic wave analysis with conventional SPT operations to enable volumetric imaging around boreholes. The method records seismic waves generated by routine SPT hammer blows at depth using a 2D geophone array deployed at the ground surface. These waveforms are then analyzed using advanced 3D full-waveform inversion techniques to reconstruct a high-resolution three-dimensional shear-wave velocity (Vs) model surrounding the borehole. Field demonstrations at sites in Florida, Kansas, and Minnesota show that the method can produce subsurface images with approximately 2-ft spatial resolution across a large 3D domain extending up to 60 ft from the borehole at any tested depth (see additional document), with Vs profiles showing strong agreement with SPT-N values.

Despite its demonstrated potential, the current implementation still requires substantial manual data processing, limiting its routine use in practice. This project aims to advance the 3D SPT-seismic method by developing automated processing workflows and practical implementation guidelines that will enable state DOTs to efficiently generate high-resolution 3D subsurface images with routine SPT site investigations. The resulting capability will help reduce subsurface uncertainty, improve risk management, and support more informed design and construction decisions for transportation projects.

The primary objectives of this research are: (1) to automate the 3D SPT-seismic method and (2) to validate its applicability across a range of geological conditions commonly encountered by State DOTs to support widespread implementation. The project will focus on developing an integrated workflow that streamlines both data acquisition and analysis, enabling the method to be deployed efficiently alongside routine geotechnical investigations without disrupting standard SPT operations.

For automation, seismic signals generated by SPT hammer blows will be continuously recorded using a surface geophone array for all blows without interfering with the drilling crew. Advanced machine-learning algorithms will be implemented to automate key data-processing tasks, including signal conditioning, noise suppression, event detection, and removal of low-quality records. The processed data will be then analyzed via advanced 3D full-waveform inversion (FWI) to reconstruct a high-resolution 3D shear-wave velocity (Vs) image around the borehole, extending up to approximately 60 ft into the surrounding subsurface.

For validation, the methodology will be demonstrated through field testing at State DOT sites, with two test locations for each participating agency, representing a variety of soil and geological conditions. The results will be compared with SPT-N values to assess accuracy and practical benefits. By automating the analysis procedures and validating the method under real field conditions, this project will help transform the SPT-seismic approach into a practical and cost-effective tool for high-resolution geotechnical site characterization.]]></description>
      <pubDate>Tue, 18 Aug 2026 16:26:47 GMT</pubDate>
      <guid>https://rip.trb.org/View/2762014</guid>
    </item>
    <item>
      <title>Evaluating Corridor Speed Management Strategies Using PeMS and Probe Vehicle Data</title>
      <link>https://rip.trb.org/View/2761174</link>
      <description><![CDATA[Utah Department of Transportation (UDOT) has recently adopted updated policies and guidance related to speed management and has begun implementing corridor treatments intended to better align vehicle speeds with roadway context. This project will evaluate how recently implemented speed management strategies affect operating speeds, travel times, and corridor traffic conditions on Utah roadways, with emphasis on strategies such as radar speed feedback signs, roadway narrowing or road diets, gateway features, landscaped medians, pavement markings, optical speed bars, rumble strips, signal timing adjustments, and other engineering treatments used to reduce excessive speeds and improve safety.

The study will use available PeMS point-detector data and ClearGuide probe vehicle data to support before-and-after evaluations across selected corridors and roadway contexts, including urban arterials, suburban corridors, and rural highways where feasible. The results will provide UDOT with practical guidance on the effectiveness of speed management treatments under Utah conditions and will produce an evaluation framework that UDOT staff can apply to future speed management projects using existing data sources. Where sufficient data are available, the study may also include a limited descriptive review of crash trends to provide additional safety context.]]></description>
      <pubDate>Sat, 15 Aug 2026 12:25:26 GMT</pubDate>
      <guid>https://rip.trb.org/View/2761174</guid>
    </item>
    <item>
      <title>AI-assisted Condition Assessment of Roads</title>
      <link>https://rip.trb.org/View/2752288</link>
      <description><![CDATA[The objective of this project is to develop an AI-assisted road monitoring system that enables low-cost, autonomous, and frequent condition-based assessments using a network of mobile sensing units. The system will use computer vision and machine learning to detect and quantify pavement defects, replacing traditional schedule-based inspections with continuous, data-driven monitoring. The proposed system provides transportation agencies with an affordable, scalable, and intelligent tool for real-time pavement monitoring. By using low-cost sensors on existing vehicles and automated data interpretation, it delivers accurate condition insights, reduces inspection costs, and supports timely maintenance decisions.]]></description>
      <pubDate>Thu, 13 Aug 2026 15:31:12 GMT</pubDate>
      <guid>https://rip.trb.org/View/2752288</guid>
    </item>
    <item>
      <title>Causally Informed Forecasting of Event Travel Dynamics for the 2028 Los Angeles Olympics Using Large-Scale Mobility Data - Phase 2</title>
      <link>https://rip.trb.org/View/2732878</link>
      <description><![CDATA[Phase 2 of this project builds on Phase 1 foundational work to develop causal inference frameworks for event-driven travel behavior, and informed by those causal insights, forecasting models for the 2028 Los Angeles Olympic and Paralympic Games. This research can directly inform the work of the White House Task Force on the 2028 Summer Olympics (established by Executive Order 14328) and broader innovations in transportation planning for large-scale events.

In Phase 1 (January–June 2026), the research team is leveraging large-scale smartphone mobility data from approximately 100,000 anonymized mobile devices to characterize travel behavior during historical major events at approximately 20 venue sites where Olympic and Paralympic Games will be held across Los Angeles County. Analyzing mobility data linked to a Points of Interest database, the team is producing descriptive analyses for two key populations: (1) event visitors, characterized by place of origin, lodging location, attendance and post-event lingering duration, and secondary activities; and (2) local residents, for whom disruptions to routine travel are analyzed before, during, and after events. From these analyses the team is constructing origin-destination (O-D) matrices disaggregated by visitor/resident status, time, and event type.

Phase 2 advances this work to develop a causal inference-informed forecasting pipeline through two interconnected research thrusts. In the first stage, the team will use quasi-experimental methods (synthetic control and difference-in differences) to estimate the causal effect of each historical event on mobility outcomes, isolating what changes in mobility were caused by the event vs. confounders including baseline trends and seasonal variation. In the second stage, an invariant prediction framework [1, 2, 3] will identify which event features reliably predict those causal effects across approximately 20 venues and diverse event types; those features then become the inputs to the forecasting model. This design is motivated by a key insight from recent work in causal transfer learning and causal representation learning: prediction performance improves when models rely on features whose relationship to the outcome remains invariant, i.e., stable across different environments, rather than based on correlational patterns that may not generalize. This causal-then-forecast approach is especially well suited to the proposed Olympic Games planning task for three reasons: there is large heterogeneity across training “environments” characterized by different venues, event scales, types, seasons, and audiences simultaneously; observations per environment are limited by modest sampling rates (1-3%) and a finite number of historical events, so restricting to causally stable features can provide performance-improving regularization; and the practical purpose is prospective planning for unprecedented scenarios (e.g., simultaneous events across multiple venues), where causally grounded models can more reliably extrapolate.

Specifically, the first thrust will employ quasi-experimental research designs, including synthetic control and difference-in-differences approaches, to estimate the causal effect of historical major events on mobility outcomes, isolating event-driven behavioral changes from baseline trends, seasonal variation, and other confounders. The second thrust uses these causal estimates to inform the development of forecasting models for predicting mobility behavior to, during, and immediately after major events at venues across Los Angeles County. In a causal transfer and representation learning framework, the causally identified event effects will be used to constrain model architectures by selecting predictive features that are invariant (i.e., that reliably predict the quantified causal effects), across event types, venues, time periods, and other sources of environmental variation. These causally informed forecasting models will generate predictions of travel demand, trip timing, O-D flows, and secondary activity patterns for both event visitors and affected local residents.

Together, these thrusts complete a research pipeline supporting scenario analysis for Olympic transportation planning spanning Phases 1 and 2: descriptive characterization of large-scale historical events, causal inference to isolate event-specific effects, and causally informed forecasting models that predict travel to, during, and after the 2028 Olympic Games. A Technical Advisory Committee (TAC) will be composed of LA Metro Olympic planning staff, mobility data providers, and Crypto.com arena data science staff, to advise on key modeling outputs and insights regarding data inputs. All data handling will comply with USC and Cuebiq privacy standards.
 
]]></description>
      <pubDate>Wed, 22 Jul 2026 17:14:39 GMT</pubDate>
      <guid>https://rip.trb.org/View/2732878</guid>
    </item>
    <item>
      <title>Inverse Problem Approaches for Bridge Structural Health Monitoring Using Displacement Data</title>
      <link>https://rip.trb.org/View/2732495</link>
      <description><![CDATA[This project aims to develop an inverse problem framework for bridge structural health monitoring (SHM) using displacement data as the primary diagnostic input. Traditional SHM methods based on finite element model updating and contact-based sensor networks are computationally demanding and require extensive field calibration, while acceleration-based techniques struggle to detect local damage. To address these limitations, the study applies inverse problem-solving methodologies that enable the direct inference of unknown structural parameters—such as stiffness variations, damage locations, and boundary conditions—from displacement measurements. Recent advancements in computer vision technologies have significantly improved the accessibility, accuracy, and cost-effectiveness of displacement data collection, making bridge condition assessment increasingly feasible. Through data analysis, inverse modeling, and validation, the research develops a validated framework for bridge condition assessment based on displacement data, reducing reliance on contact sensor networks and improving the accuracy of local damage detection across transportation infrastructure.]]></description>
      <pubDate>Wed, 22 Jul 2026 12:08:48 GMT</pubDate>
      <guid>https://rip.trb.org/View/2732495</guid>
    </item>
  </channel>
</rss>