Home Skip to main content

T&L Faculty and Students Present at INFORMS Conference

Posted: Nov 7, 2019

NDSU Transportation and Logistics faculty and students presented their research at the 2019 INFORMS Annual Conference Oct. 20-23 in Seattle. INFORMS is the world's largest professional association dedicated to and promoting best practices and advances in operations research, management science and analytics to improve operation processes, decision-making, and outcomes. The annual conference draws more than 7,200 members, students, academic and industry experts.

  • Bukola Bakare, transportation and logistics Ph.D. student, presented, "Does Being Socially Responsible Correlate with Traffic Congestion in Metropolitan Areas in the United States?" Traffic congestion is a critical environmental event with a direct adverse impact on people's health and productivity. This study investigates the level of socially responsible Fortune 500 corporations in major cities in the United States. Through interviews conducted by the authors, qualitative analysis is used to study one of the top metropolitan areas to capture and explore how traffic congestion trends and corporate social responsibility ratings apply on a local level. Joseph Szmerekovsky, chair of NDSU's Department of Transportation, Logistics and Finance, was a coauthor.
  • Bhavana Bhardwaj, transportation and logistics Ph.D. student, presented "Railroad Track Irregularities Position Accuracy Assessments Using Low Cost Sensors. Bhardwaj's research investigates the potential use of low-cost sensors aboard hi-rail vehicles to monitor automatically and continuously for inertial events caused by irregular track geometry. Due to the GPS receivers position error, this study introduces a signal processing and statistical method to estimate the position of peak inertial events from multiple traversals and validates its accuracy by comparing the estimated positions of detected irregularities with the actual positions of irregularities recorded by railroad inspector.
  • Raj Bridgelall, assistant professor of transportation and logistics and researcher with the Upper Great Plains Transportation Institute, presented "Scholarly Teaching in Transportation Science and Technology" in a session focused on improving scholarly teaching and Moderated by Joseph Szmerekovsky, chair of the NDSU Department of Transportation, Logistics and Finance. Bridgelall noted that transportation systems have been undergoing a rapid transformation integrating advancements in sensing, artificial intelligence, and communications technologies. The motivations behind this transformation can be classified into push and pull factors that drive business gains and address societal needs, respectively. In the context of scholarly teaching, Bridgelall developed a new push-pull-impedance framework to motivate and organize learning about factors that affect the evolution and adoption of smart transportation. He provided an example of using the framework to demonstrate learning about the potential societal impacts and implications of connected vehicle technology.
  • Neeraj Dhingra, transportation and logistics Ph.D. student, presented "Text Mining Railroad Accident Narratives for Identifying Contributors to Railroad Accidents and to Extend the Sustainability of Fixed Field Data." Dhingra used advanced data mining techniques to examine historical data about railroad accidents, particularly text descriptions, to identify key factors that caused the accident and to enhance information from fixed-field data. The techniques could help identify prevention strategies.
  • Sajad Ebrahimi, transportation and logistics Ph.D. student, presented "A Hybrid Framework to Identify and Prioritize the Most Important Supply Chain Enablers to Improve Sustainability." In the study, Ebrahimi aimed to identify and prioritize influential enablers to reach a sustainable supply chain using a combination of advanced statistical analysis and structural modeling. Szmerekovsky was a coauthor.
  • S. Ali Haji Esmaeili, transportation and logistics Ph.D. student presented "An Optimization Approach to the Market Incentive Analysis of Second-generation Bioethanol Supply Chains. In his paper Esmaeili notes that first generation bioethanol is produced from food-based biomass which has raised social issues such as controversy of diverting production from food to fuel. His study examines the use of monetary incentives to promote second-generation bioethanol which is produced from non-edible biomass feedstocks. To analyze the impacts of incentives, Esmaeili compared the supply chains of first-generation and second-generation bioethanol production. Joseph Szmerekovsky, chair of NDSU's Department of Transportation, Logistics and Finance, and Ahmad Sobhani, assistant professor of management information systems at Oakland University in Rochester, MI, were coauthors.
  • Mingwei Guo, transportation and logistics Ph.D. student, presented "Spatial Analysis in the Decision Making of CEP Crowd-sourced Last-mile Delivery." Soaring E-commerce sales and other last-mile delivery needs have put pressure on traditional delivery modes. Crowd-sourced last-mile delivery can play a role in carrier, express and parcel (CEP) services and impact the location selection of warehouses and package hubs. The research used spatial analysis tools to examine cases of CEP last-mile needs and provide reference for further study.

    Guo also presented "The Use of Hybrid Crowd-Sourcing Models in Last-mile Delivery of Carrier, Express and Parcel (CEP) Services. Third party logistics (3PL) services have been growing rapidly as a result of online shopping. This dynamic business environment has exerted huge pressure on CEP services. The research used crowd-sourcing logistics tools to work with CEP last-mile delivery to help solve the logistics problems posed by occasions of peak demand. The researchers examined the potential reasons and causes which will trigger the use of crowd-sourcing logistics investigated potential partnerships that would improve the entire CEP logistics system. Szmerekovsky was a coauthor.
  • Phat Huynh, Ph.D. student in industrial and manufacturing engineering, presented "Probabilistic Graphical Model of Acute Disorder Pathogenesis for Patient-specific Preventive Treatment." Disease pathogenesis has not been encoded in machine-learning models because of its complex temporal dependencies and inter-patient variability. Huynh proposes a pathogenesis probabilistic graphical model (PPGM) that incorporates pathogenetic domain knowledge to capture mechanisms leading to disease onsets. The model was evaluated by two case studies: obstructive sleep apnea and paroxysmal atrial fibrillation.
  • Narendra Malalgoda, transportation and logistics Ph.D. student, presented "Do Transportation Network Companies Reduce Public Transit Use in the U.S." The study examines the effect of transportation network companies (TNCs) like Uber and Lyft and transit effectiveness (reliability, availability, etc.) on public transit ridership in the U.S. The study found that transit effectiveness of both bus and rail transit declined; TNC availability significantly increased rail transit ridership in 2014 and 2015, but the effect subsided in 2016 and 2017; and transit effectiveness was highly significant for rail transit, and when examining its effect year-by-year, rail transit effectiveness trumped TNCs availability. Coauthor was Siew Hoon Lim.
  • Ali Rahim-Taleqani, transportation and logistics Ph.D. student, presented "Location Prediction Using Recurrent Neural Network." Given the increasing volume of data from dockless scooter/bike programs, the more difficult it is for managers to predict where users will travel and use bikes/scooters. Taleqani used a recurrent neural networks (RNN) model to analyze the spatial and temporal scooter/bike data, showing significant improvements over similar methods.
  • Fangzheng Yuan, transportation and logistics Ph.D. student presented, "A Patient No-show Predictive Model with Limited Data and Information for Outpatient Appointments." In the study Yuan developed a decision tree model for predicting the no-show probability of outpatient appointments for a medical center and compared it with some commonly used models. This study demonstrates an alternative way to predict patient's no-show probability based on limited data and information. Szmerekovsky was a coauthor.
NDSU Dept 2880P.O. Box 6050Fargo, ND 58108-6050