Informs Annual Meeting 2017

WA60

INFORMS Houston – 2017

WA58

combination carriers should assist their customers in positioning air cargo capability as part of their integrated supply chain strategies. Such effort will help air cargo avoid commoditization. 2 - An Airport Task Assignment Problem with Travel Time and Purity Consideration Fan Xiao, Ms, Tongji University, Zhangwu Rd 1, Tongji Building A, 1913, Shanghai, 200092, China, 892225786@qq.com, Zhe Liang We study an airport task assignment problem with travel time and purity consideration, which is to assign a list of airport tasks to a set of shifts designed for airport personnel, so that the total operation cost is minimized. There are two major differences between the problem studied in this paper and the traditional assignment problem. First, we consider the travel time between different tasks, so it is not only to determine the assignment of the tasks, but also the routing between different tasks. The second is to consider shift purity in this problem, i.e., to limit the number of task types/qualifications in each shift. We propose a column generation algorithm to solve the problem efficiently. 3 - Flight Outsourcing to Regional Airlines Soheil Sibdari, Associate Professor, University of Massachusetts Dartmouth, 1933 Commonwealth Avenue, Unit 408, Boston, MA, 02135, United States, ssibdari@umassd.edu, Farbod Farhadi In this paper we address the reasons behind outsourcing flights to regional airlines. We show how outsourcing improves airline operations by adjusting their capacity. We also study the competitive effects of flight outsourcing on the U.S. domestic airline market. 4 - Sustainable Hub Location Model under Uncertainty in Aviation Industry Suresh Kumar Jakhar, Assistant Professor, Indian Institute of Management Lucknow, Lucknow, 226013, India, skj@iiml.ac.in, Himanshu Rathore, Devendra Singh Literature on Sustainable Hub Location models (SHLP) is limited because of their computational complexity and multidimensional nature (Mohammadi et al. 2014). We propose a realistic SHLP model in aviation industry by using chance constraint programming nested in queuing theory (M/D/s) to incorporate noise pollution and GHG emissions during the entire landing and take-off cycle of the aircraft. It determines a sustainable network strategy. To solve the NP hard formulation, we propose a methodology based on Fireworks metaheuristic algorithm embedded in Benders decomposition approach. 5 - A Stochastic Optimization Model for Fleet Assignment under Uncertainty Conditions

362E Facility Location I Sponsored: TSL, Facility Logistics Sponsored Session Chair: Trevor S Hale, University of Houston-Downtown, One Main Street, Houston, TX, 77002, United States, halet@uhd.edu 1 - Dynamic Ambulance Location with Time-dependent Parameters: A Two-stage Stochastic Programming Chun Peng, PhD Candidate, Beijing Institute of Technology, Haidian District, 5 South Street, Beijing, 100081, China, pengchun12.18@163.com Chun Peng, PhD Candidate, HEC Montreal, Montreal, QC, H3T.2A7, Canada, pengchun12.18@163.com In this paper, we highlight a novel two-stage stochastic programming framework that accounts for stochastic nature of emergency demand and time-dependent parameters for multi-period ambulance location. Besides minimizing total cost, we embed chance constraint in the recourse to guarantee a reliable coverage level. Furthermore, we present another stochastic model with generalized chance constraint that forcing violated probability over all possible level, to make a trade- off between coverage and cost management. An enhanced decomposition scheme is developed for the large-scale MIPs. Finally, a real-life case study is conducted and some managerial insights are drawn for EMS managers. 2 - Mathematical Programming Approaches for Facility Location Problems with Neighborhoods Victor Blanco, University of Granada, Dpt. Quant. Meth. Economics & Business, Fac. Ciencias Economicas y Empresariales, Granada, 18011, Spain, vblanco@ugr.es We introduce a new version of the classical p-median facility location problem in which it is assumed that the exact position of the potential facilities is unknown. Instead, assuming that the facilities are embedded in $\mathbb{R}^d$, the open facilities are required to belong to a region around their initial estimated positions: the neighborhoods. Two decisions have to be made: (1) determinate the open facilities; and (2) compute of the positions of them into their neighborhoods, at global minimum cost. We present several MINLP formulations for ordered median location problems and describe two math-heuristic approaches that allow us to solve larger instances at smaller CPU times. 3 - Population-based Metaheuristics for Fairer Facility Location Christopher M. Rump, Applied Statistics and Operations Research, Bowling Green State University, 344 Business Administration Building, Bowling Green, OH, 43403-0267, United States, cmrump@bgsu.edu, Cameron R.King, Robert Green We study complements to classic facility location problems on a discrete network. These complementary problems provide fairer solutions for decision makers seeking to consider the impact beyond extremal focus on either one (center) or all (median) customers. However, solving these problems is more computationally demanding than their well-known counterparts as they lose the incentive that customers should travel to the closest open facility. We employ two meta- heuristics, particle swarm optimization and adaptive large neighborhood search, to find good solutions to these intractable problems. 4 - A Graph Theoretic Approach to the Location Allocation Problem on a Network Trevor S.Hale, Professor, University of Houston - Downtown, One Main Street, Houston, TX, 77002, United States, halet@uhd.edu, Ryan Pepper, Michelle Hutnik, Faizul Huq Building upon previous research that developed graph-theoretic solutions for the location-allocation problem on a path, this research constructs graph-theoretic solutions for the location-allocation problem on cycles, stars, and complete graphs. This knowledge is then leveraged to help predict solutions for more general graphs. Some avenues of future research are also delineated. 362F Aviation Applications Contributed Session Chair: Daniel Felipe Otero, Universidad de los Andes, Bogotá, Cundinamarca, Colombia, otero.df@uniandes.edu.co 1 - Airline Cargo Business Efficiency Measurement using DEA_PCA Hossein Najmi, University of North Texas, Denton, TX, 76203- 5017, United States, hossein.najmi@unt.edu, Seock-Jin Hong In this study, we show that air cargo growth rate is declining. In addition, a negative relationship exists between the cargo business ratio and airline’s efficiency in airlines, which serve passengers and cargo (combination carriers) using exclusive freighter from 2008 to 2015. The authors suggest that WA59

Daniel Felipe Otero, Instructor, Universidad de los Andes, Cr 1 este #19 A.40, Bogotá, Cundinamarca, Colombia, otero.df@uniandes.edu.co, Edwin Camilo Quiroga Gomez, Andres Medaglia

The variability in operational times has a great impact on aircraft scheduling. This work analyzes fleet assignment, considering the stochastic behavior of operational times. To solve the Fleet Assignment Problem (FAP), we propose a stabilized column generation problem, applying a set of specialized auxiliary shortest path models, based on lineal and dynamic programming (k-Shortest Path Problem with Resource Constraint - SPPRC). Finally, employing a Monte Carlo simulation, the On - Time Performance (OTP) of the fleet assignment is evaluated in specific scenarios.

WA60

370A Approaches to Teaching Operations Management Sponsored: INFORMEd Sponsored Session Chair: Dorothy Klotz, Fordham Univeristy, 140 West 62nd Street, New York City, NY, 10023, United States, klotz@fordham.edu 1 - Building the Complete Operations Management Course Around a Web-based Simulation Game Ardavan Asef-Vaziri, Professor, College of Business and Economics, Dept of Systems and Operations MGMT, 18111 Nordhoff Street, Northridge, CA, 91330-8378, United States, aa2035@csun.edu We describe how a simulation game can be implemented as a bridge interfacing the concepts, techniques, and procedures thought in an OM course and real- world applications. Students examine their theoretical take-away in a pseudo-real-life environment. Our experience shows that after playing these games, the students realize that understanding the course material plays a significant role in their performance in real-life applications.

463