2016 INFORMS Annual Meeting Program

TA11

INFORMS Nashville – 2016

TA11 104A-MCC Novel Applications of Network Optimization Sponsored: Optimization, Network Optimization Sponsored Session Chair: Anna Nagurney, John F. Smith Memorial Professor, Isenberg School of Management, University of Massachuseets, Amherst, MA, 01003, United States, nagurney@isenberg.umass.edu 1 - A General Multitiered Supply Chain Network Model Of Quality Competition With Suppliers Dong Li, Arkansas State University, dli@astate.edu, Anna B Nagurney In this paper, we develop a general multitiered supply chain network equilibrium model consisting of competing suppliers and competing firms who purchase components from suppliers. The competitive behavior of each tier of decision- makers is described along with their strategic variables, which include quality of the components and, in the case of the firms, the quality of the assembly process itself. The governing equilibrium conditions of the supply chain network are formulated as a variational inequality and qualitative properties are presented. The algorithm, accompanied with convergence results, is then applied to numerical supply chain network examples along with sensitivity analysis. 2 - Supply Chain Network Equilibrium With Strategic Financial Hedging Using Futures Contracts firms’ procurement activities are exposed to several risk factors such as commodity price uncertainty and exchange rate volatility. The firms can use futures contracts to hedge the risks. Our research studies the equilibrium of the entire network where each firm optimizes its own operation and hedging decisions. 3 - A Time-space Network Model For Medical Resources Allocation In An Epidemic Outbreak Ding Zhang, SUNY Oswego, New York, NY, United States, ding.zhang@oswego.edu This paper presents a dynamic logistics model for medical resources allocation that can be used in the control of an epidemic diffusion. It couples a forecasting mechanism, constructed for the demand of a medicine in the course of such epidemic diffusion, and a logistic planning system to satisfy the forecasted demand and minimize the total cost. The model is built as a closed-loop cycle, comprises of forecast phase, planning phase, execution phase, and adjustment phase. The parameters of the forecast mechanism are adjusted in reflection of the real data collected in the execution phase by solving a quadratic programming problem. A numerical example is presented to illustrate the model. 4 - A Generalized Nash Equilibrium Network Model For Post-disaster Humanitarian Relief Anna Nagurney, John F. Smith Memorial Professor, University of Massachusetts Amherst, Amherst, MA, 01003, United States, nagurney@isenberg.umass.edu, Emilio Alvarez Flores, Ceren Soylu We develop a Generalized Nash Equilibrium network model for post-disaster humanitarian relief by nongovernmental organizations (NGOs). NGOs derive utility from providing relief supplies to the victims of the disasters at multiple demand points in a supply chain context while competing with each other for financial funds provided by donations. The shared constraints consist of the lower and upper bounds for demand for relief items at the demand points and can be imposed by the regulatory body or higher level coordinating NGO to reduce materiel convergence and congestion. We provide an effective computational scheme and numerical examples plus solutions under the Nash Equilibrium counterpart. Zugang Liu, Pennsylvania State University - Hazleton, Hazleton, PA, United States, zxl23@psu.edu, Jia Wang We develop a modeling and computational framework for supply chain networks with strategic financial hedging. We consider multiple competing firms that purchase multiple materials to manufacture their products. The supply chain

TA12 104B-MCC

Mixed-Integer Programming with Applications Sponsored: Optimization, Integer and Discrete Optimization Sponsored Session Chair: Minjiao Zhang, Kennesaw State University, 560 Parliament Garden Way, Kennesaw, GA, 30144, United States, mzhang16@kennesaw.edu 1 - Coordinated Capacitated Lot-sizing For Multiple Product Families With Setup Times Tiffany Bayley, University of Waterloo, Waterloo, ON, Canada, tiffany.bayley@uwaterloo.ca, Haldun Sural, James H Bookbinder We examine a coordinated capacitated lot-sizing problem for multiple product families, where demand is deterministic and time-varying. The problem considers only holding costs, where capacity constraints limit the number of item and family setups and the amount of production in each period. Using a strong reformulation and relaxing the demand constraint, we improve both the upper and lower bounds using Benders decomposition and a cutting plane procedure. Through computational experiments, we show that our method consistently achieves better bounds, reducing the duality gap compared to other single-family methods studied in the literature. 2 - A Polyhedral Study On Lot-sizing Problem With Capacity Acquisition Jia Guo, The University of Alabama, Tuscaloosa, AL, 35487, United States, jguo23@crimson.ua.edu, Minjiao Zhang Determining the optimal capacity level to invest is one of the fundamental problems for an enterprise’s operation. In this study, we consider a single-echelon lot-sizing problem with capacity acquisition (CALS). Two families of the so-called capacity definition and demand satisfying inequalities are proposed to describe the convex hull of CALS. Our computational results show that the proposed inequalities are very effective in solving CALS. 3 - The Slim Branch And Price Method With An Application To The Hamiltonian P-median Problem Ahmed Marzouk, Texas A&M University, College Station, TX, 77840, United States, ambadr@email.tamu.edu, Erick Moreno-Centeno, Halit Uster We present the Slim Branch and Price (SBP) method which is an improvement over traditional Branch and Price in the case of binary master problems. The main advantage in SBP is that the branching tree has only one main branch with several leaves. We illustrate the computational advantage of SBP over Branch and Price on the Hamiltonian p-median problem. In particular, under one hour limit, SBP can solve to optimality instances with up to 200 nodes; whereas Branch and

Price can solve to optimality instances with up to 127 nodes. 4 - Multiechelon Lot Sizing With Intermediate Demands

Ming Zhao, Assistant Professor, University of Houston, Houston, TX, 77204, United States, mzhao@bauer.uh.edu, Minjiao Zhang We prove that multiechelon lot sizing with intermediate demands is NP-hard. However, in the case of fixed number of echelons, we are able to derive polynomial time algorithms for both capacitated and uncapacitated models.

TA13 104C-MCC Computational Linear Optimization General Session

Chair: Nikolaos Ploskas, Carnegie Mellon University, 5000 Forbes Avenue, Pittsburgh, PA, 15213, United States, ploskasn@gmail.com 1 - Vector Space Decomposition For Solving Linear Programs Marco Lübbecke, RWTH Aachen University, marco.luebbecke@rwth-aachen.de, Jean Bertrand Gauthier, Jacques Desrosiers We propose a general algorithmic framework for solving linear programs. An iteration moves from one solution to the next in a direction with a certain step size. Part of the direction is determined by a pricing problem that is interesting in its primal and dual form: the dual fixes part of the dual variables and optimizes the rest; the primal selects a convex combinations of variables. Several known algorithms are unified by our view, in particular the primal simplex method, the minimum mean cycle canceling algorithm, and the improved primal simplex method. The framework allows us to precisely characterize when degenerate iterations can occur, and how to avoid them (of course, at a computational cost).

236