Project 1- Estimation of logistic transportation system performance under extreme weather condition: A data-driven approach
Research Priority: Improving Mobility of People and Goods
Principal Investigator(s): Yunlong Zhang (TAMU), Lee Han (UTK), Bruce Wang (TAMU)
Project Partners: Oak Ridge National Lab
Project Description: This is the year two effort of the 2-year project. The objective is to utilize a data-driven approach (e.g., Machine Learning and Deep Learning) to robustly, proactively, and trustworthily estimate the impact of impending extreme weather events (e.g., tropical cyclones) on key logistical infrastructure elements, including ports and highways. Beyond the first year’s research, we plan to broaden our objectives from focusing on the port system (multiple points) to encompassing the interconnected network of the highway system. The highway network will be conceptualized as a graph in which intersections and roads are nodes and edges, respectively. A Graph Neural Network (GNN) equipped with temporal attention will be utilized to estimate the impact of tropical cyclones on highway operational performance, leveraging its ability to process complex spatial and temporal dependencies within the highway network.
Project 2 – Generating reliable freight disruption measures with freight telematics data (Year 2)
Research Priority: Improving Mobility of People and Goods
Principal Investigator(s): Salvador Hernandez (OSU), Lee Han (UTK), Steven Jiang (NCAT)
Project Partners: EROAD, Robinsight
Project Description: In the aftermath of disasters that challenge the resilience of transportation networks, the urgency for planning for rapid mobility and recovery has been underscored. The primary objective of resilience is to enable transportation agencies to prepare more effectively for such events. In this light, resilience measures serve as a critical tool, providing a means to assess the impact of disruptions and inform strategic investments to mitigate these occurrences. The first year of our research addressed the critical challenges faced by states and agencies in measuring freight network systems: the scarcity of comprehensive data and the inadequacy of analytical methods. While there is substantial data available on the movement of people and passenger vehicles, understanding freight movements—especially under disruptive scenarios—poses distinct challenges. Freight movements, governed by corporate supply chain decisions, are subject to constant change due to various economic conditions and span multiple jurisdictions and transport modes. Moreover, methods to capture and analyze data that encompasses these complex dynamics have been limited. With a focus on these challenges, our initial research presented a novel framework that leveraged Robinsight telematics data to bridge this gap. In the first year, we have delved into the telematics data to explore its capacity for developing robust freight network resiliency measures, with the trucking sector in the Tennessee and Pacific Northwest.
Project 3 – A Micromobility Framework for Freight Deliveries in Urban Environments
Research Priority: Improving Mobility of People and Goods
Principal Investigator(s): David Porter (OSU), Salvador Hernandez (OSU), Philip Carleton (OSU)
Project Partners: E-cargo bike manufacturers and companies such as B-Line and Stuart
Project Description: Micromobility is broadly defined as “any small, low-speed, human- or electric-powered transportation device, including bicycles, scooters, electric-assist bicycles, electric scooters (e-scooters), and other small, lightweight, wheeled conveyances.” (Shaheen et al., 2015; FHWA, 2021). In recent years, micromobility transportation options such as cargo cycles (including manually- and electrically-powered bicycles and tricycles) have been considered by many cities as a potential alternative for making last-leg deliveries (Ding et al., 2024; Narayanan & Antoniou, 2022; Rudolph & Gruber, 2017; Tipagornwong & Figliozzi, 2014). These advantages include their ability to use both roadways and cycle lanes and their potential to reduce shipping fees (since they save in parking, congestion charges, and fuel costs).
Project 4 – Shipper’s utility functions according to commodity groups and freight modal split
Research Priority: Improving Mobility of People and Goods
Principal Investigator(s): Bruce Wang (TAMU), Lee Han (UTK), Yunlong Zhang (TAMU)
Project Description: The U.S. freight transportation network consists of railroads, highways, waterways, pipelines, and airways. It carries the flow of commodities along the supply chains from the origin of production to the destination of consumption. The already congested transportation system faces fast-growing freight demands. The annual freight volume is expected to grow from 19 billion tons in 2022 to at least 29 billion tons in 2050, according to the US Bureau of Statistics. Imports and exports account for over 13% of the total freight volume and will grow at a much faster rate than domestic freight. The mode and route choices of shippers will play an important role in network performance and in the planning consideration of planners and public policy makers because they determine the freight volumes on the network according to their modes, and therefore determine congestion and delay on the routes. From a microscopic and operational perspective, to address bottlenecks and choke points on the network requires understanding of the shipper’s choice of mode and routes. Shippers make their rational choices based on the premise of maximizing their utility functions. In this project, we proposed to study the utility functions and other means for route and mode choices by the shippers. However, shipper’s choices also have to do with the commodity types. For example, higher value commodities such as machinery or electronic products may be less sensitive to the price charged than bulk commodities. The goal of the project is to propose a framework for predicting the network commodity flows so that policy makers and planners may use. The midterm goal of the effort is twofold. The first is to explore the format of the shipper’s utility function in their decision of mode and route choices. The Second is to study the differences between the utility functions of different commodity groups by using a few groups that have data available. This project focuses on exploring the shippers’ utility function: the format and calibration by using example commodity groups specifically.
Project 5 – Impact of Multimodal Freight Network on Private Sector Global Distribution (Phase 2)
Research Priority: Improving Mobility of People and Goods
Principal Investigator(s): Hector Vergara (OSU), Salvador Hernandez (OSU)
Project Partners: EROAD
Project Description: Multimodal freight transportation enables a more efficient use of transportation infrastructure by allowing freight to switch between at least two different modes of transport (e.g., truck, rail, water/marine, and air) from the point of origin to the point of destination [1], [2]. The multimodal network affects private sector distribution system planning and operations as the physical infrastructure and policies of the managing organizations impose constraints that are relevant for strategic, tactical, and operational level decisions. At the strategic level, private sector companies develop and adapt their distribution systems through the location of manufacturing plants, distribution centers, and warehouses, largely based on the established multimodal network infrastructure. Accordingly, the private sector’s distribution system efficiency depends on the inherent resiliency and efficiency of the multimodal network [1], [3].
Project 6 – Embedded Hybrid Truck-Drone Delivery Routing Design for Rural Areas
Research Priority: Improving Mobility of People and Goods
Principal Investigator(s): Yang Zhou (TAMU), Bruce Wang (TAMU)
Project Description: According to the U.S. Department of Agriculture, around 14% of the U.S. population lives in rural areas, with a higher percentage of seniors and increased poverty rates compared to urban areas. Rural regions face unique challenges such as sparse population density, limited infrastructure, and geographic isolation, leading to economic and social disparities. To address these issues, logistics companies like DHL and UPS have introduced hybrid truck-drone delivery systems for better service coverage. This hybrid system combines the long-range capabilities of trucks with the flexibility of drones for last-mile delivery, offering potential solutions for rural logistic challenges. However, the success of these systems depends on their efficiency, particularly in solving the truck- drone routing problem to optimize delivery routes.
Project 7 – Delivery Deserts: Mapping, Understanding, and Overcoming Service Challenges
Research Priority: Improving Mobility of People and Goods
Principal Investigator(s): Marcella Kaplan (UTK), Kevin Heaslip (UTK)
Project Description: The evolution of delivery services has reshaped consumer behavior and expectations for availability, speed, and cost. This transformation has not been uniform, leading to a divide that impacts society. The delivery/logistics industry has not adopted business models to mitigate inequities in delivery services, creating “delivery deserts”—regions where accessibility and efficiency of delivery services are severely limited or non-existent. This research addresses the limited exploration of delivery deserts, a concept that diverges from the commonly studied “food deserts.” These deserts, often found in rural areas affects the backbone of e-commerce and access to essential goods like groceries and medicines.
Project 8 – Corridor Planning Tool for Efficient and Resilient Agricultural Supply Chain in California
Research Priority: Improving Mobility of People and Goods
Principal Investigator(s): Shailesh Chandra (CSLB)
Project Partners: Caltrans
Project Description: In California, which leads in agricultural output and its economy relies on an extensive freight network, there exists a gap in providing stakeholders with the necessary guidance to assess the resilience of critical infrastructure elements such as corridors, intermodal connectors, and aging bridges for efficient movement of agricultural products and its supply chain. The objective of this proposed research is to enable stakeholders to identify and implement strategies that would improve the efficiency and resilience of freight networks for the movement of key agricultural commodities in California as well as in other states in which agricultural supply chain constitutes a significant part of the economy. The research approach will involve developing a framework for risk priority of various elements of agricultural freight infrastructure in California.
Project 9 – Risk-Informed Stochastic Programming with Applications to Inland Flash Flooding of Roadway Freight Networks
Research Priority: Improving Mobility of People and Goods
Principal Investigator(s): Mingzhou Jin (UTK)
Project Partners: TDOT
Project Description: From a practical perspective, annual US freight volume is expected to grow by at least 50% in the next 25 years (US Bureau of Statistics) [1], putting even more pressure on the already-strained road network system. Ensuring supply chain resilience through freight mobility in the face of disruptions is critical for national security and the economy (White House Report 14017, 2021) [2]. The proposed research will give stakeholders a tool to identify strategic investment opportunities for enhancing roadway resilience. The proposed research will fill two main gaps in the literature. First, transportation resilience research has nearly exclusively focused on the impacts on coastal regions through sea level rise. The proposed research will explicitly focus on the inland impact namely from flooding caused by more frequent high-intensity storms through both a short- and long-term lens. Second, current decision-making frameworks struggle in the presence of low-probability, high-impact events (such as flash flooding), either by ignoring these events and suffering the full brunt of their impact or by overemphasizing them and wasting resources. Our proposed research exploits the benefits of each approach by biasing our scenario generation toward extreme events, providing a more complete understanding of system vulnerabilities and opportunities for strategic investment without increasing model size too much. The approach can be applied to any long-tail distribution and can account for stakeholder risk postures.