Informs Annual Meeting Phoenix 2018

INFORMS Phoenix – 2018

TB30

PanelistS Dries Goossens, Ghent University, Tweekerkenstraat 2, Gent, 9000, Belgium Stefan Creemers, IESEG France, Lille, France Margret V. Bjarnadottir, Breidavik 87, Reykjavik, 112, Iceland Julie Ann Williams, University of West Florida, Pensacola, FL, United States n TB28 North Bldg 221A Joint session RAS/TSL: Intermodal Sponsored: Railway Applications Sponsored Session Chair: Mike D. Prince, BNSF Railway, Fort Worth, TX, 76244, United States Co-Chair: Justin Goodson, Saint Louis University, St. Louis, MO, 63108, United States 1 - Opportunities for Optimization and Automation in Intermodal Terminal Operations Steven Jay Tyber, General Electric, Chicago, IL, 60613, United States The recent surge in intermodal volumes and truck driver shortage has pushed many rail terminals to their limits. In this talk, we survey the challenges faced by intermodal terminals and discuss the central role optimization and automation will play in driving operational efficiency and improved utilization of limited physical real-estate. 2 - The Role Of Partial Information And Commitment In Dynamic Transportation Procurement Pol Boada-Collado, Northwestern University, Evanston, IL, United States, Karen Smilowitz, Sunil Chopra We analyze decision processes related to commitment contracts in transportation procurement. We consider a novel setting motivated by industry with short-term demand visibility, adding additional dynamism. We model the decision process of selecting transportation modes when partial information is available and show how partial information fundamentally changes contracting policies. 3 - Investigating the Capacity of Different Intermodal Terminal Layouts with AnyLogic Wesley Chen, University of Illinois Urbana-Champaign, 205 N. Mathews, Graduate Center, Urbana, IL, 61801, United States Freight railways in the United States are currently transporting record volumes of containers and trailers in intermodal service. An efficient and reliable network of intermodal trains requires inland intermodal transloading facilities with sufficient capacity to handle growing traffic volumes. In this research, AnyLogic simulation is used to examine different terminal layouts and operating factors and their relationship to the overall capacity of an intermodal terminal. 4 - Appointment Scheduling for Intermodal Dray Operations Lisa Tang, Schneider, Green Bay, WI, 54313, United States Effective Appointment Setting (AS) is a key influencer in the success metrics of intermodal operations. While the primary objective of successful AS is to facilitate productive routing options for the subsequent driver dispatch process, decisions made at this level also impact other profit measures, such as driver capability for new order acceptance and ramp dwell. In this session, we will discuss Schneider’s network model approach to providing optimal appointment suggestions, which takes into account shipment forecast, empty miles, ramp dwell and driver schedules. We will also review how model output can be used to provide feedback on driver capability. n TB29 North Bldg 221B Mobility on Demand Systems with Transit Integration Sponsored: TSL/Urban Transportation Sponsored Session Chair: Samitha Samaranayake, Cornell University, Ithaca, NY, 14853, United States 1 - On-demand Multimodal Transit Systems: A Case Study Pascal Van Hentenryck, University of Michigan, 1813 IOE Building, 1205 Beal Avenue, Ann Arbor, MI, 48108-2117, United States This talk describes the design and operations of an on-demand multimodal transit systems, combining different fleets of vehicles. A case study for the city of Ann Arbor is presented and offers key insights on a potential future for transit agencies.

2 - On the Interaction between Autonomous Mobility-on-Demand and Public Transportation Systems Marco Pavone, Stanford Unviersity, Stanford, CA, United States In this talk I will discuss models and algorithms to integrate autonomous mobility-on-demand systems with public transportation. In particular, I will discuss the design of socially-optimal control policies, present a real-world case study for New York City, and discuss a number of opportunities for future research. 3 - Operational Challenges of using On-Demand Services as a First/Last Mile Solution – Some Insights from Ford Crystal Wang, Ford Motor Company, Detroit, MI, United States Some city agencies are interested in integrating On-Demand service for the first/last mile solution and Mass-Transit for the rest of the trip. However, this solution does not work for all cities. We will present the simulation results of On- Demand first/last mile solution for different city types. On the other hand, Ford implemented On-Demand service for first/last mile solution in Shanghai in 2016. We will talk about some interesting observations and real challenges from this pilot, and encourage researchers for further investigation. 4 - Integrating Mobility-on-demand Services with Mass Transit Samitha Samaranayake, Cornell University, School of Civil & Environmental Engineering, 220 Hollister Hall, Ithaca, NY, 14853, United States While ride-hailing services provide a valuable service, as evident by their popularity, there are many questions regarding their scalability, efficiency, equity and externalities (e.g. congestion, pollution etc.). For these systems to be a comprehensive mobility solution, they need to be better integrated with higher capacity services such as mass-transit. Operationalizing such a system requires the ability to efficiently match riders across multiple services in real time, a computationally challenging task to solve at an urban scale. We present some models for operating such a multi-modal on-demand service and discuss some of the computational challenges including hardness results. n TB30 North Bldg 221C Joint Session Smart Cities/MIF: Energy Aspects Towards a Smart City Emerging Topic: Smart Cities Emerging Topic Session Chair: Krystel Castillo, University of Texas-San Antonio, San Antonio, TX, 78249, United States 1 - Optimal Sizing and Stochastic Control of a Micro-Grid Using Behind-the-meter Battery Storage and Photovoltaic Panels Ying Chen, PhD, UTSA - Texas Sustainable Energy Research Institute, San Antonio, TX, 78249, United States, Krystel Castillo, Bing Dong We present an efficient building energy management system installed with a behind-the-meter battery energy storage (BES) system and photovoltaic panels. In this micro-grid system, BES system as a buffer is used to alleviate the stochastic nature from the solar energy and dynamic short-term demands from the building. In this study, we use wait-and-see idea from stochastic programming with the martigale model of forecast evolution method to achieve the averaged optimal battery size and power trading limit for this system. And then, we use myopic and lookahead policies with deterministic forecasts and stochastic forecasts to implement the control for this sequence problem, respectively. 2 - A Simulation Approach to Allocating Microgrid Generation for Resilience Against Extreme Weather Tongdan Jin, Texas State University, 601 University Drive, Ingram School of Engineering, San Marcos, TX, 78666, United States, Priyadharshini Kumaravelan, Honggang Wang Microgrid has the capability of enhancing grid resilience through defensive operations in contingency. We present a simulation model for designing resilient distribution network via distributed wind and solar integration. We strive to achieve three performance goals manifested as prevention, survivability and recovery. 3 - Blockchain-Assisted Crowdsourced Energy Systems Ahmad F. Taha, PhD, The University of Texas at San Antonio, San Antonio, TX, 78249, United States Crowdsourcing relies on people’s contributions to meet certain objectives. This work explores a framework for Crowdsourced Energy Systems (CES), where small-scale energy generation or energy trading is crowdsourced from distributed energy resources. An operational model for CESs with different types of crowdsourcees is proposed. The model yields market equilibrium depicting distributed generaiton setpoints. Crowdsourcing incentives are designed to steer crowdsourcees to the equilibrium. As the number of crowdsourcees/transactions scales up, a secure energy trading platform is required. To that end, the presented research is integrated with a lightweight blockchain implementation.

285