Informs Annual Meeting Phoenix 2018

INFORMS Phoenix – 2018

MC27

4 - Service Systems with Rationally Inattentive Customers Caner Canyakmaz, Postdoctoral Researcher, ESMT European School of Management and Technology, Schlossplatz 1, Berlin, 10178, Germany, Tamer Boyaci Many service systems are partially visible such that customers are not always able to discern precise queue lengths upon arrival. This stems mainly from potential information frictions due to environment and/or cognitive capabilities of customers. We analyze rational queueing behavior under limited customer attention through rational inattention framework. Customers optimize their information acquisition process trading-off the benefits of better information against its cost. We characterize equilibrium strategies and investigate the effect of information cost on throughput and social revenue. n MC27 North Bldg 132B Student Attitudes Towards Learning OR/MS/Analytics Sponsored: Education (INFORMED) Sponsored Session Chair: Susan Wright Palocsay, James Madison University, Harrisonburg, VA, 22807, United States 1 - Moderator Susan Wright Palocsay, James Madison University, Msc 0202, Computer Info Sys & Bus Analytics Dept, Harrisonburg, VA, 22807, United States Analytics is providing new opportunities to create business value. Consequently, coverage of quantitative analysis/modeling skills has increasing importance in undergraduate education. The panel will discuss the role of students’ attitudes in learning OR/MS/analytics. Panelists Natalie M. Scala, Towson University, Dept. of e-Business and Technology Management, 8000 York Road, Towson, MD, 21252, United States 1 - Academic Motivation in Introductory Business Analytics Courses Baback Vaziri, James Madison University, 2210 Reserve Circle, Unit 203, Harrisonburg, VA, 22801, United States, Stacey Kelly, Elham Torabi, Luis J. Novoa The purpose of this study is to examine relationships among student perceptions of their learning experience, including levels of motivation and effort, and their performance in introductory business analytics courses using the MUSIC Model of Motivation. The results from this study will be used to identify research-based motivation strategies that can be used to improve instruction and suggest techniques to improve student learning experiences. Specifically: 1) students’ rating of model components, 2) strategies to improve model component scores, and 3) predictive or explanatory trends in model component scores with effort and performance. n MC28 North Bldg 221A Train Design and Service Planning Sponsored: Railway Applications Sponsored Session Chair: Seyed Mohamad Nourbakhsh, BNSF 1 - Simulating Railcar Transit under Different Operating Strategies Tzu-Yu Chang, University of Illinois at Urbana Champaign, Urbana, IL, United States, Darkhan Mussanov This research investigates how railway classification yard operating strategies and train schedule flexibility affect the average railcar transit time across a network. Simulation results suggest that, for all the three operating strategies studied, transit time increases as the level of schedule flexibility increases. Depending on the specific traffic and blocking scenario, different operating strategies result in the shortest average railcar transit time. The resulting trends may help railway practitioners in formulating more robust train and yard operating plans. 2 - An Integrated Train Service Plan Optimization Model with Variable Demand Xuesong Zhou, Arizona State University, Tempe, AZ, United States, Lingyun Meng A well designed train timetable should fully utilize the limited resources to maximize operators’ profits and passenger travel demand satisfaction. This talk explains an integrated demand/service/resource optimization model for managing the above-mentioned three key decision elements. By using a Lagrangian relaxation solution framework to recognize the dual costs of both passenger travel demand and limited resources of track and rolling stock, we decompose the formulation into a novel team-based train service search sub-problem for

maximizing the profit of operators. Numerical experiments are conducted to examine the effectiveness of the dual and primal solution search algorithms. 3 - Optimization Models for Block Re-design Problem Chinmoy Mohapatra, BNSF Railways, Arlington, TX, United States, Anantaram Balakrishnan Railroads develop blocking plans to consolidate shipments into groups (or blocks) at intermediate yards so as to minimize the total handling costs and transit miles of all shipments. We model the blocking problem as a network design problem with maximum degree and node flow constraints, and consider restrictions based on material types. We discuss modeling and algorithmic enhancements to effectively solve this large-scale integer program, and present computational results for actual problem instances. 4 - Trade-off Between Efficiency and Equity under Time Dependent Passenger Demand in Railway Timetabling Dewei Li, Beijing Jiaotong University, Beijing, China, Tianyu Zhang, Xinlei Dong The aim of this paper is to analyze and to improve the efficiency and transportation equity of urban railway timetable. A MIP model is proposed in which the train operation constraints and train capacity constraints are taken into account , an adaptive large neighbor algorithm is designed to solve this problem. The feasibility of the model and algorithm is verified by the Changping line of Beijing Metro. A sensitivity analysis on the weight of efficiency and equity in objective function illustrates that different weight can cause different result in the aspect equity and efficiency. 5 - Visualizing the Effects of Maintenance on Train and Yard Performance Trefor Williams, Rutgers University, Piscataway, NJ, United States, John F. Betak Using actual data from a railroad in the United States, graph network visualizations have been constructed to better understand the linkages between the location of maintenance activities such as rail replacement and surfacing with delays to through freights and missed switching deliveries to customers. This analysis also includes the use of text mining to identify typical maintenance problems, major problems and accident locations from the railroads maintenance records. We will demonstrate how these visualizations can be used to identify critical areas in the railroad’s track network where delays are frequently caused by maintenance activities. n MC29 North Bldg 221B Innovations in Last-Mile Urban Package Delivery Sponsored: TSL/Urban Transportation Sponsored Session Chair: Bo Zou, University of Illinois, IL, 60607, United State 1 - Same-day Delivery Using In-store Customer with Store Transfers Alp Arslan, Rotterdam School of Management, Pannekoekstraat 58C, Rotterdam, 3011LJ, Netherlands, Niels Agatz This paper considers a system in which in-store customers can make deliveries on their way to home. They may deliver directly to the home of an online customer or transfer the package at another store after which another in-store customer makes the final delivery. To examine the potential of such store transfers, we present an optimization approach to match delivery tasks and in-store customers in real-time. 2 - Integrated Modeling of Crowdsourced Urban Delivery with Rebalancing Consideration Bo Zou, University of Illinois, 2073 Engineering Research Facility, 842 W. Taylor Street, Chicago, IL, 60607, United States This research investigates using crowdshipping for urban delivery. We focus on a specific type of crowdshipping that deals with delivering shipments from spatially distributed locations such as restaurants, and retail, grocery, and drug stores to customers who are also spatially distributed, within a guaranteed time. We consider not only assigning shipments dynamically to compatible crowdsourcees to allow for time-guaranteed delivery, but also strategically repositioning crowdsourcees at the same time so that crowdsourcee supply can be better matched spatially with shipment demands. 3 - Designing Electronic Marketplaces for Transportation Services Arim Park, Rutgers University, Newark, NJ, 07102, United States, Yao Zhao, Soohyun Cho, Seongtae Kim On-demand matching services for commercial transportation needs have only recently entered the market in South Korea, with early players like Uber Freight and Convoy seeking new ways to solve inefficiencies in the transportation sector. With this in mind, we would like to design an efficient mechanism for facilitating the growth of an electronic market of transportation services (trucking) in Korea. To this end, we would first conduct empirical studies to identify the preferences of both customers and carriers (truckload freight carriers). We provide meaningful guidance in the matching process between shippers and carriers in e-marketplace.

194