2016 INFORMS Annual Meeting Program

WE59

INFORMS Nashville – 2016

WE59 Cumberland 1- Omni Supply Chain Networks Design under Uncertainty Sponsored: TSL, Facility Logistics Sponsored Session Chair: Nima Sadghiani, University of Michigan, Ann Arbor, MI, 12121, United States, nsalehi@umich.edu 1 - Rolling Stock Maintenance Location Routing: Stochastic And Robust Programming Formulations Denise Désirée Tönissen, Eindhoven University of Technology, Eindhoven, Netherlands, D.D.Tonissen@tue.nl, Zuo-Jun Max Shen, Joachim Jacob Arts The maintenance location routing problem for train units is a NP-hard problem, where we locate maintenance facilities, while also taking the maintenance routing into account. Facility location is a long term strategic decision, but the optimal facility locations depend on line planning and rolling stock schedules. Since these change on a regular basis, our objective is to minimize the costs under different discrete line planning and rolling stock schedule scenarios. We give robust and stochastic formulations, provide algorithms to solve the problems and perform numerical tests. 2 - Global Sourcing Under Correlated Uncertainties: A Scenario-based Stochastic Programming Approach Nima Salehi Sadghiani, University of Michigan, 3545 GreenBrier Managing uncertainty is one of the challenging issues in supply chains. Using only the expected value of the uncertain parameters for sourcing decisions in a deterministic model can be risky due to the uncertainties that threaten both the optimality and feasibility of the decision variables. Most previous studies on supply chain network design assume that uncertainties are independent. In this study, we address the need for an effective tool to incorporate potential uncertainties associated with the model parameters and their correlation into sourcing decisions. WE60 Cumberland 2- Omni Collaborative Logistics Sponsored: TSL, Freight Transportation & Logistics Sponsored Session Chair: Brett Shields, UTSI, 411 B.H. Goethert Parkway, Tullahoma, TN, 1, United States, bshields@utsi.edu 1 - A Classification For Collaborative Logistics State Of The Art Analysis Andrea Rusich, Univeristy of Trieste, Andrea.Rusich@phd.units.it Scientific literature on collaborative logistics still misses a harmonized approach to analyse its management implications. This paper aims to fill this gap by introducing a classification canvas based on four perspectives: decision-makers, form of collaboration, ICT and decision technology enablers, operations management. The proposed analysis enhances the understanding of existing collaborative logistics approaches and identifies further research steps. In addition, an application example to the Physical Internet shows how the classification could support the study of emerging collaborative logistics frameworks. 2 - Vehicle Routing Problem In Collaborative Environments: Introducing An Exchange Mechanism For Orders. Vincent Karels, TUE, v.c.g.karels@tue.nl Considering the distribution of orders, we introduce a system that exchanges orders in such a way that cooperating distribution companies maintain their competitiveness and autonomy while also decreasing their costs. First an instrument is formulated which determines the marginal costs of any order within the vehicle routing problem, while retaining that the calculations remain computationally tractable. Subsequently an auctioning mechanism utilizing this instrument is introduced, and the measure to which this mechanism evolves to Blvd,, Apt. 39A, Ann Arbor, MI, 48105, United States, nsalehi@umich.edu, Mark Stephen Daskin, Don Zhang, Mark Everson, Michael Sanders the best allocation of orders (the social optimum) is evaluated. 3 - Fair Sharing Approach For Network-wide Adoption Of Collaborative Container Logistics Rob A Zuidwijk, Professor of Ports in Global Networks, Erasmus University Rotterdam, P.O. Box 1738, Rotterdam, 3000 DR, Netherlands, rzuidwijk@rsm.nl, Irina Romochkina, Peter van Baalen This paper draws from cooperative game theory while studying fair sharing schemes and is inspired by actual issues in the adoption of Inter-Organizational Systems (IOS) in container logistics in a sea port environment. We demonstrate

(1) the influence that network effect, as experienced in business network communication structure and propensity for coordination between players, has on the chances of an IOS to be fully adopted by a business network under fair sharing conditions, and (2) that a fair sharing scheme has the potential to eliminate two types of conflicts related to uneven benefits distribution and through this enhance IOS adoption under certain conditions.” 4 - A Stochastic Make Or Buy Model Considering Additive Manufacturing In Supply Chain Brett Shields, University of Tennessee Knoxville, 411 B.H. Goethert Parkway, Tullahoma, TN, United States, bshields@utsi.edu, Javad Seif, Andrew Yu In this project a two-stage stochastic program is presented in a make or buy setting, considering the supply of parts produced by additive manufacturing. This model generalizes a number of works in the literature and provides a framework for implementing additive manufacturing in supply chain during times of uncertainty.

WE61 Cumberland 3- Omni Facility Location Contributed Session

Chair: Zhaomiao Guo, Universtiy of California, Davis, 609 Anderson Road,, Apt. 264, Davis, CA, 95616, United States, zmguo@ucdavis.edu 1 - Service Network Design With Equilibrium Constraints Based On General M/g/1 Waiting Time Mahdi Hamzeei, Post Doctoral Fellow, University of Maryland, Baltimore, 685 West Baltimore Street, MSTF 7-00D, Baltimore, MD, 21201, United States, hamzeei.m@gmail.com, Amir Ahmadi Javid We consider the service network design in which an operator determines some facilities and their size to open. The customers then choose facilities to send their demand such that their time is minimized. Customers’ time consists of a travel and a waiting time. Assuming M/G/1 waiting time in each facility, using conjugate duality and decomposition approaches, we propose and efficient method to solve this problem. This structure was motivated by several applications such as preventive healthcare and disaster management. We conclude by presenting preliminary numerical results. 2 - Service Level Restrictions In Location Problems With Disproportionate Assignment Costs Jeremy W North, Murray State University, 9400 Wickliffe Rd., Wickliffe, KY, 42087, United States, jnorth@murraystate.edu, Robert M Nauss A common assumption in facility location modeling is that customer assignment cost is proportional to distance. In this work, we drop this assumption and analyze the subsequent effect on service level when weighted average distance is minimized. Computational experiments contrast the p-median with an augmented version enforcing higher overall service levels. We show that our modeling approach is more economical than implementing a closest assignment restriction. Additionally, a Greedy Random Adaptive Search Procedures (GRASP) algorithm is applied to solve larger problem instances. 3 - A Continuum Approximation Method For Reliable Facility Location Design With Imperfect Information Hongqiang Fan, University of South Florida, Tampa, FL, United States, hongqiangfan@mail.usf.edu, Xiaopeng Li, Lifen Yun In this paper, we propose a continuum approximation method to solve the reliable facility location design problem with imperfect information. Then a discrete algorithm is proposed to obtain discrete facility locations. The results of case studies indicate that this model also can get the near-optimal solution compared with the discrete model and has a robust performance. 4 - Facility Location In A Competitive Market: Incorporating Network Congestion Effects Zhaomiao Guo, Universtiy of California, Davis, 609 Anderson Road,, Apt. 264, Davis, CA, 95616, United States, zmguo@ucdavis.edu, Yueyue Fan We assess facility networks in a competitive market, where the infrastructure system is shaped by collective actions of multiple decision entities from both supply and demand sides. A network-based multi-agent optimization modeling framework is developed to reflect “selfish” behaviors of individual service investors and users and to simultaneously capture the interactions among all over a network structure. To overcome computational difficulty imposed by non- convexity of the problem, we rely on recent theoretical development on variational convergence of bivariate functions to design a solution algorithm with analysis on its convergence properties.

502