Informs Annual Meeting 2017

TC61

INFORMS Houston – 2017

2 - Investigating the Relationship Between Volume, Blocking Complexity and Railway Classification Yard Performance with Simulation Tyler Dick, University of Illinois at Urbana-Champaign, 1241 Newmark Lab MC-250, 205 N Mathews Avenue, Urbana, IL, 61801, United States, ctdick@illinois.edu, Nao Nishio In designing a train plan for a railway network, one decision is to determine the number of blocks to be handled by each train and the number of blocks to be assembled at each classification yard while maintaining a certain level of service. This research uses simulation and factorial experiment design to investigate the relationship between yard throughput volume, the total number of blocks processed and the level of service. Following this initial analysis, the research will be expanded to examine the influence of train arrival rate, overall yard configuration and other yard operational factors on the developed capacity relationship. 3 - Yard Planner: A Digital Force Multiplier for Railroad Classification Yards Jeremiah Dirnberger, Product Manager, GE Transportation, Jacksonville, FL, United States, jeremiah.dirnberger@ge.com GE’s Yard Planner, part of a broader “Yard of the Future” vision, is a decision support system designed to aid the planning of activities in classification yards. It provides real time planning and optimization for all car processing activities, such as train-to-track assignments, block-to-track assignment, build sequence, etc. The pilot is being developed for hump yards with Norfolk Southern, but efforts are already under way at GE to expand the product to support flat-switching operations of varying size and complexity. This presentation will discuss the solution approach, lessons learned through the development process, and planned/actual results from the pilot roll-out at Macon Yard. 370A Addressing Challenges in Teaching Analytics Sponsored: INFORMEd Sponsored Session Chair: Janet M. Wagner, Stockton College of New Jersey, 101 Vera King Farris Drive, Galloway, NJ, 08205, United States, janet.wagner@stockton.edu 1 - A Process Mining Approach Towards Identification of Recurring Behavioral Patterns in Course Sequences Jie Tao, Assistant Professor in Information Systems, Fairfield University, 1073 N.Benson Rd, Fairfield, CT, 06824, United States, jtao@fairfield.edu Course planning at universities have received increasing interests from both researchers and practitioners. In this article, I propose an analytical framework in order to identify recurring course behavioral patterns from course sequencing datasets. The framework processes and analyzes the raw course sequencing datasets for the purpose of presenting insightful, actionable rules. A case study is then conducted in order to demonstrate and evaluate the proposed framework, with the course sequencing data of MBA students at XYZ University. After empirical evaluations , the findings from the case study are correct and actionable, in both contexts of course sequencing/planning and process mining. 2 - Teaching Students to Frame the Analytics Problem Satish V. Nargundkar, Clinical Professor, Georgia State University, 2802 Fairlane Dr., Atlanta, GA, 30340, United States, snargundkar@gsu.edu, Subhashish Samaddar Students often have difficulty interpreting a business case and framing it as an analytics problem when needed. Specifically, learning to identify a dependent variable, and determining if a continuous or binary dependent is more appropriate is the first step. Identifying an outcome period and a time frame from which to collect sample data is the next task. Finally, what questions should one ask of the client before the project begins? Examples are presented to help students understand these critical tasks at the start of an analytics project. 3 - Insights to the Usefulness of More Gamification in Classrooms Sinan Tas, University of Wisconsin-Platteville, 1 University Plaza, Platteville, WI, 53818, United States, tass@uwplatt.edu, Pamela Tas Gamification is using game elements in a classroom setting in order to make classes more engaging for an effective learning experience. Many educators may find it challenging to figure out how much to gamify their classes. Is there too little or too much gamification? In this empirical study, we compare different levels of gamification to determine if there are significant differences in student learning and motivation. TC60

4 - Specification Grading in an Introductory Business Analytics Course Janet M.Wagner, Dean, Stockton University, 101 Vera King Farris Drive, Galloway, NJ, 08205, United States, janet.wagner@stockton.edu Specification grading is an alternative to common “points based” grading systems, and its proponents claim it supports course rigor, increases student motivation, while simultaneously reducing the instructor’s grading time. In this talk I will discuss how I set up a core undergraduate course in Business Analytics (in the Information Systems Curriculum) around specification grading, and discuss how well this system did or didn’t support the three goals above. Spoiler alert: it worked pretty well.

TC61

370B Sports Analytics III

Sponsored: SpORts Sponsored Session Chair: Benjamin Grannan, Virginia Military Institute, Lexington, VA, 24450, United States, grannanbc@vmi.edu 1 - Optimized Team Scheduling for Collegiate Baseball Summer Leagues Benjamin Grannan, Assistant Professor, Furman University, Greenville, SC, 29601, United States, grannanb@gmail.com An optimization based approach to team scheduling for collegiate summer baseball leagues is presented. 2 - Airline Revenue Management for Mega Sporting Events Yuqi Peng, University of South Carolina, Columbia, SC, United States, Yuqi.Peng@grad.moore.sc.edu, Mark Ferguson, Ovunc Yilmaz Using game information from the 2012 NCAA Men’s Basketball Championship and an extensive database of air travel ticketing data, we explore the connection between the knockout game results (Sweet Sixteen and Elite Eight) and air travel demand for the neutral site where the next stage (Final Four) will be played. We use a difference-in-differences approach to identify specific demand patterns and present market-level heterogeneity. Based on our findings, we offer a new pricing strategy to help the airline industry improve their revenue management solutions around sporting events. 3 - Round-robin Tournaments Generated with the Circle Method Have Maximum Carry-over Dries Goossens, Ghent University, Tweekerkenstraat 2, Gent, 9000, Belgium, Dries.Goossens@ugent.be, Frits C.Spieksma, Annette Ficker, Erik Lambrechts The Circle Method (aka the polygon method, or the canonical procedure) is widely used in the field of sport scheduling to generate schedules for round-robin tournaments. The so-called carry-over effect value is a number that can be associated to each roundrobin schedule; it represents a degree of balance and fairness of a schedule. We prove that, for an even number of teams, the Circle Method generates a schedule with maximum carry-over effect value, answering an open question. 4 - Predicting MLB Game Outcomes through Simulation Justin Long, Slippery Rock University, 5902 Westridge Circle NW, North Canton, OH, 44720, United States, jbl1005@sru.edu In this study, Major League Baseball is explained using a Markov Chain Method, as each game state is independent of all previous states excluding the one immediately prior. To model this, data from every game in the 2010 through 2015 seasons are used to calculate probabilities of all possible state transitions. Offensive playing ability and home-field advantage are taken into consideration by partitioning probabilities into different transition matrices. For a given team, simulations are utilized to mimic the series of an entire season; this result provides an overall expected win-percentage. Predictions can be used by managers, players, and fans for varying purposes.

357