2016 INFORMS Annual Meeting Program

WA88

INFORMS Nashville – 2016

WA86 GIbson Board Room-Omni Telecommunications Modeling and Analysis Sponsored: Telecommunications Sponsored Session Chair: Dimitri Papadimitriou, Nokia Bell Labs, Antwerp, Belgium, Belgium, dimitri.papadimitriou@nokia.com 1 - Reducing The Internet Adoption Gap Between Rich And Poor Through Auction Mechanisms Sergio Cabrales, Universidad de los Andes, Bogota, Colombia, s-cabral@uniandes.edu.co, Luis Andrés Marentes, Yezid Donoso, Tilman Wolf, Anna B Nagurney The latest Millennium Development Goals Report by the United Nations has found that poor populations are behind in their internet adoption due to high prices relative to available budgets. We design an auction mechanisms to suit the allocation of bandwidth to user needs and their budgets. The article develops another dimension to the topic of dynamic pricing models design, which is resource allocations to favor target groups by finding Nash Equilibria of underlying games using extreme value theory and a self-discrimination induced on users. Results indicate the auction mechanism let increase allocation for the population being part of the target group during peak periods. 2 - An Efficient Sampling-based Algorithm For Chance-constrained Two-stage Problems Jianqiang Cheng, Sandia National Laboratories, Livermore, CA, United States, jianqiang.cheng@gmail.com, Richard Li-Yang Chen We consider a chance constrained version of two-stage stochastic optimization problems which minimizes the sum of the first-stage costs and the $p$-quantile of the second-stage random costs. To solve this problem, we first apply sampling- based approximation techniques, precisely, the partial sample average approximation, to obtain an approximate deterministic formulation. Then, we develop decomposition algorithms to solve the approximation problems. Computational results on a stochastic network design show the strength of our proposed approximation approach. 3 - On The Convex Piecewise Linear Unsplittable Multicommodity Flow Problem Bernard Fortz, Université Libre de Bruxelles, Brussels, Belgium, bernard.fortz@ulb.ac.be, Luis Gouveia, Martim Joyce-Moniz We consider the problem of finding the cheapest routing for a set of commodities over a directed graph, such that: i) each commodity flows through a single path, ii) the routing cost of each arc is given by a convex piecewise linear function of the load i.e. the total flow) traversing it. We propose a new mixed-integer programming formulation for this problem. The linear relaxation of this formulation gives an optimal solution for the single commodity case, and produces very tight linear programming bounds for the multi-commodity case. We also derive new valid inequalities for the compact basic model based on the projection of the extended formulation. 4 - Mixed-integer Programming Model For The Joint Function Placement And Assignment Problem Dimitri Papadimitriou, Bell Labs, dimitri.papadimitriou@alcatel-lucent.com Function-oriented networks take as input demands described by unsplittable finite sequences of operations and perform by executing at each node at most one out of the n possible operations part of the sequence. The problem consists of selecting the subset of nodes where to jointly place function operators and assigning demands to paths crossing these nodes without exceeding both their processing and arc capacity. Following the objective of minimizing the sum of location, allocation and routing cost, we formulate the corresponding mixed- integer program. Numerical experiments are conducted to evaluate the performance tradeoffs with different placement and routing schemes/constraints.

WA87 Broadway A-Omni Production and Scheduling Contributed Session

Chair: Rasaratnam Logendran, Oregon State University, School of Mech lndustrial & Mfgr Engr, Rogers Hall Rm 204, Corvallis, OR, 97331- 6001, United States, logen.logendran@oregonstate.edu 1 - A Scheduling Algorithm For Additive Manufacturing Kai-Oliver Zander, PhD Student, Texas Tech University, Box 43061, Lubbock, TX, 79409-3061, United States, Kai-Oliver.Zander@ttu.edu, Milton Louis Smith Recent studies have shown that additive manufacturing (AM) can enable an increase in efficiency and generate an enhanced customer value. The increased utilization of AM will lead necessarily to practical problems regarding production scheduling. This presentation introduces a new scheduling method specifically designed for an AM production environment with multiple machines. Based on existing research, a new algorithm has been developed to allow an efficient scheduling and batching of jobs for AM machinery. A simulation study shows the effectiveness of the developed algorithm. 2 - Hybird Robust And Stochastic Production Planning On The Shop Floor Considering Real Time Information Assembly and fabrication factories are universally challenged with the need to continually reduce costs and improve efficiency while simultaneously becoming increasingly flexible to meet ever-changing customer demand. A hybrid decision making model is proposed to address the uncertainties on the shop floor considering real time information. Stochastic programming is adopted to deal with unexpected machine breakdown. Robust optimization is utilized to address the demand uncertainty considering the worst-case scenario. The goal is to minimize the total production cost and the worst-case cost associated with unmet demand. A case study based on a manufacturing shop floor is presented. 3 - Quantifying The Performance Of The Tabu Search/Path Relinking Algorithm For Batch Scheduling In Hybrid Flow Shops Rasaratnam Logendran, Professor, Oregon State University, School of Mech lndustrial & Mfgr Engr, Rogers Hall Rm 204, Corvallis, OR, 97331-6001, United States, logen.logendran@oregonstate.edu, Omid Shahvari We address a batching and scheduling problem in hybrid flow shops with the objective of simultaneously minimizing total weighted completion time and total weighted tardiness. It is assumed that dynamic job release and machine availability times exist, batch sizes can have desired lower bounds, and jobs can skip one or more stages. The performance of the tabu search/path relinking algorithm is evaluated based on tight lower bounds identified by the column generation technique. Zhengyang Hu, Research Assistant, Iowa State University, 100 Enrollment Services Center Ames, Ames, IA, 50011, United States, zhengya@iastate.edu, Guiping Hu

WA88 Broadway B-Omni Military Applications I Contributed Session

Chair: Ali Pala, PhD Student, University at Buffalo, SUNY, 271 Palmdale Drive, Apt 5, Buffalo, NY, 14221, United States, alipala@buffalo.edu 1 - Military Modeling Of Unconventional Conflict Dean S Hartley, Principal, Hartley Consulting, 106 Windsong Lane, Oak Ridge, TN, 37830, United States, DSHartley3@comcast.net Unconventional conflict refers to conflicts involving at least one nation state and which is not dominated by conventional combat. There is significant overlap between unconventional conflict and operations other than war (OOTW) and irregular warfare (IW). While unconventional conflict needs a whole of government approach, the military is the only organization that is organized and staffed to undertake large and long-term modeling efforts in this domain. This presentation will investigate many of the issues in modeling unconventional conflict.

393