Informs Annual Meeting 2017

SA61

INFORMS Houston – 2017

SA61

2 - Integrating Series in the Simultaneous Allocation of Slots on a Network of Airports Paola Pellegrini, IFSTTAR, Villeneuve d’Ascq, France, paola.pellegrini@ifsttar.fr, Tatjana Boli , Lorenzo Castelli, Raffaele Pesenti Airport capacity of many airports is managed through the airport slots. SOSTA is an integer linear programming formulation for the simultaneous allocation of slots at all European airports, considering aircraft turnaround. It reproduces the current process by bringing optimization into it, improving the resulting slot allocation. The original formulation however cannot be used to allocate slots for the entire scheduling season since it does not consider the allocation of a series of slots. In this work, we study possible approaches for overcoming this limitation. 3 - Modelling Efficiency, Fairness and Acceptability in Airport Slot Scheduling Decisions Jamie Fairbrother, Centre for Transport and Logistics (CENTRAL), Lancaster University Management School, Fylde College, Mathematics and Statistics, Lancaster, LA1 4YW, United Kingdom, j.fairbrother@lancaster.ac.uk, Konstantinos G. Zografos We propose and solve a slot scheduling model which optimizes efficiency and considers fairness and airline acceptability. We introduce a new congestion based fairness measure and we allow airlines to express their preferences regarding the satisfaction of their slot requests. We apply the proposed model to a medium size congested airport to assess the impact of existing and modified IATA slot scheduling guidelines on fairness and efficiency. 4 - Multimarket Contact and Capacity: Evidence from the United States Airline Industry Hao Su, University of Maryland-College Park, Robert H. Smith School of Business, College Park, MD, 20742, United States, haosu@rhsmith.umd.edu, Martin E.Dresner This paper will examine the relationship between multimarket contact (MMC) and airline capacity. Based on the literature, we hypothesize that: (1) MMC negatively affects capacity and (2) MMC has a more negative effect on a carrier’s capacity when that market is not important to the carrier. Based on U.S. air carrier traffic data with carefully defined markets, we hypothesize that MMC will be associated with lower route capacity and that this effect is greater on less important routes. Our findings may provide evidence for the mutual forbearance hypothesis, suggesting that MMC may facilitate tacit collusion on the supply of capacity. 370A Case Competition I Sponsored: INFORMEd Sponsored Session Chair: Palaniappa Krishnan, University of Delaware, Newark, DE, 19716-2130, United States, baba@udel.edu 1 - Staffing at the Strategic Advisory Consulting Group: the “S” Word Elliott Weiss, University of Virginia, Charlottesville, VA, United States, WeissE@Darden.virginia.edu Staffing at the Strategic Advisory Consulting Group: The “S” Word” is a field- based case that requires deterministic and stochastic modeling for a staffing problem. The case can be used to reinforce quantitative topics related to simulation analysis, newsvendor thinking and optimization. Managerial judgment and creativity are required with respect to implementation of solutions. The case is intended for use in introductory courses in OM and DA/QA or in elective courses in Management of Service Operations or Optimization. It can be used to cover optimization techniques and applied modeling. 2 - Glitz Investments Srinivas Krishnamoorthy, Simon Fraser University, Vancouver. BC, Canada, srinivas_krishnamoorthy@sfu.ca, Shelly Bajaj Rylee Smith - an investment manager - wishes to identify the factors that explain successful movies in Bollywood. She uses log linear and quantile regres- sion to arrive at her conclusions. The case can be used in an elective course in a BBA or MBA program or a core course in a MS in Analytics program. SA60

370B MIF Paper Competition Sponsored: Minority Issues Sponsored Session Chair: Lauren Berrings Davis, North Carolina A&T State University, Greensboro, NC, 27411, United States, lbdavis@ncat.edu Co-Chair: Sean Barnes, University of Maryland-College Park, College Park, MD, 20742, United States, sbarnes@rhsmith.umd.edu 1 - Minority Issues Forum Award Karen T.Hicklin, University of North Carolina at Chapel Hill, B-24 Hanes Hall, Chapel Hill, NC, 27599-3260, United States, khicklin@email.unc.edu The annual MIF Paper Competition features paper submissions covering an array of topics from members of the MIF community. The prize committee evaluated submissions based on importance of the topic, appropriateness of the approach, and impact of the paper within its respective area. Four finalists have been selected to present their research: Alp Akcay and Canan Corlu (Simulation of inventory systems with unknown input models: A data driven approach), Michelle Alvarado and Lewis Ntaimo (Chemotherapy appointment scheduling under uncertainty using mean-risk stochastic integer programming), Esra Buyukhtahtakin and Joseph C. Hartman (A mixed-integer programming approach to the parallel replacement problem under technological change) and Maria Mayorga, Emmett Lodree, and Justin Wolczynski (The optimal assignment of spontaneous volunteers). 2 - Simulation of Inventory Systems with Unknown Demand Models Canan Corlu, Eindhoven University of Technology, Eindhoven, Netherlands, na, Alp Akcay Assuming the availability of limited demand data, we consider the stochastic simulation of an inventory system with unknown demand distributions. Building on a nonparametric Bayesian approach, we propose a simulation replication algorithm that estimates a service level for the inventory system without making any assumptions on the functional form of the demand models. We illustrate our approach in a single-product inventory simulation. 3 - Chemotherapy Appointment Scheduling under Uncertainty using Mean-Risk Stochastic Integer Programming Lewis Ntaimo, Texas A&M.University, 3131 TAMU, College Station, TX, 77843, United States, ntaimo@tamu.edu, Michelle Alvarado Chemotherapy involves a series of appointments over several weeks and the timing of these appointments is critical to the patient’s healing process. However, oncology clinics often find it very challenging to schedule large volumes of cancer patients for chemotherapy due to limited resources and data uncertainties such as nurse availability, patient acuity levels and appointment durations. In this work, we develop risk-averse stochastic integer programming models and algorithms towards solving this problem. Computational results based on real data show that the new approach significantly reduces patient waiting times and nurse overtime. 4 - A Mixed-Integer Programming Approach and New Cutting Planes for the Parallel Replacement Problem under Technological Change Esra Buyuktahtakin, New Jersey Institute of Technology, Newark, NJ, United States, esratoy@njit.edu, Joseph Hatrman The parallel replacement problem under economies of scale (PRES) determines minimum-cost keep-and-replace decisions for a group of assets. We study the MIP formulation of PRES under technological change (PRES-T), such that newer, technologically advanced assets have higher capacity than assets purchased earlier. We provide optimal solution characteristics and insights into the economics of the problem and derive new cutting planes for optimizing it. Computational experiments on the USPS fleet management case illustrate that Maria Esther Mayorga, North Carolina State University, 400 Daniels Hall, Dept. of Industrial & Systems Engineering, Raleigh, NC, 27695, United States, memayorg@ncsu.edu, Emmett Lodree, Justin Wolczynski In this presentation, a queuing model is formulated to accurately represent the dynamics of assigning spontaneous volunteers to tasks (or jobs) in a post-disaster setting. In particular, we model volunteers as servers with not only stochastic arrivals but also with stochastic abandonment times. We formulate a continuous- time Markov decision process and find an optimal policy for volunteer allocation under the relatively stable recovery phase of post-disaster management. We then test the optimal policy against heuristic policies using a simulation model. We also test the performance of these policies under relaxed assumptions. the inequalities are quite effective in solving PRES-T instances. 5 - The Optimal Assignment of Spontaneous Volunteers

23