EURO|INFORMS 26TH EUROPEAN CONFERENCE ON OPERATIONAL RESEARCH

MMXIIIRoME

ROME 1-4 JULY, 2013

ABSTRACT BOOK

TECHNICAL SESSIONS

Monday, 8:30-10:00

� MA-02Monday, 8:30-10:00O1-2

Discrete Optimal Control, Multi-objectiveControl of Discrete Processes andDynamic Positional Games

Stream: Discrete Optimal ControlInvited sessionChair: Gerhard-Wilhelm Weber, Institute of Applied Mathematics,Middle East Technical University, ODTÜ, 06531, Ankara, Turkey,gweber@metu.edu.trChair: Vladimir Zubov, Mechanics of continuum media,Institution of Russian Academy of Sciences DorodnicynComputing Centre of RAS, Vavilova 40, 119333, Moscow,Russian Federation, zubov@ccas.ru

1 - Application of OR to the Control of Metal Crystalliza-tion Process in CastingAlla Albu, Applied Optimization Problems, Institution ofRussian Academy of Sciences Dorodnicyn Computing Centreof RAS, Vavilova 40, 119333, Moscow, Russian Federation,alla.albu@mail.ru, Vladimir Zubov

The process of metal crystallization in casting is considered. To ob-tain a model of a good quality it is desirable the shape of solidificationfront to be close to a plane and its law of motion to be close to a pre-set one. To find a regime of metal solidification that will satisfy thetechnological requirements the optimal control problem is posed. Thespeed at which the mold moves is used as the control. The optimalcontrol problem was solved numerically using gradient methods. Thiswork was supported by RFBR N12-01-00572-a and by the Programfor Fundamental Research of Presidium of RAS P18.

2 - Determining the solutions for dynamic pozitionalgames with informational extended strategies.Boris Hancu, Faculty of Mathematics and Informatics,Moldova State University, Mateevici str., 60, MD-2009,Chisinau, Moldova, Moldova, Republic Of,boris.hancu@gmail.com

We study the non informational extended games which are generatedby the two directional flow informational extended strategies of theplayers. The theorem about the existence of the Nash equilibriumprofiles in this type of the games is proved. The static game withinformational extended strategies can be treated as a dynamic gamewith incomplete information and non informational extended strate-gies, where nature makes the first move, but not everyone observesnature’s move. A parallel algorithm for determining the optimal strate-gies in this dynamic game is elaborated

3 - Investigation of the Fast Automatic DifferentiationTechniqueVladimir Zubov, Mechanics of continuum media, Institutionof Russian Academy of Sciences Dorodnicyn ComputingCentre of RAS, Vavilova 40, 119333, Moscow, RussianFederation, zubov@ccas.ru, Alla Albu

The generalized FAD-methodology allows to obtain the precise valueof the functional gradient. However, the discrete conjugate problemmay not approximate the continuous conjugate problem, thus the dis-crete gradient wouldn’t approximate the gradient of the continuousproblem. The consequence of this can be a convergence to a discreteoptimal control, which doesn’t approximate the continuous optimalcontrol. It is proposed to use approximations of the cost functionaland differential constraints that will allow to avoid this problem. Thework was supported by RFBR N12-01-00572-a.

� MA-03Monday, 8:30-10:00O1-3

Managing service systems

Stream: MSOM Service Management SIG StreamInvited sessionChair: Jan Van Mieghem, Kellogg School of Management,Northwestern University, 2001 Sheridan Road, Jacobs Center, 5thFloor, 60208, Evanston, IL, United States,vanmieghem@kellogg.northwestern.edu

1 - Server Scheduling in Large Overloaded QueueingSystems with Impatient CustomersNahum Shimkin, Electrical Enginnering, Technion, Israel,shimkin@ee.technion.ac.il

We consider an overloaded multi-class queueing system with multiplehomogeneous servers and impatient customers. Customers of differentclasses are assigned to servers with the goal of minimizing the wait-ing and reneging costs. For the case of exponential patience and linearwaiting costs, a certain index rule (the c-mu/theta rule) is optimal inan asymptotic, fluid-limit sense. In this talk we provide several exten-sions of the asymptotic model, that include non-linear waiting costs,general waiting distributions, and constraints on the waiting times inthe different queues.

2 - Does the Past Predict the Future? The Case of DelayAnnouncements in Service SystemsRouba Ibrahim, University College London, WC1E 6BT,London, United Kingdom, rouba.ibrahim@ucl.ac.uk, MorArmony, Achal Bassamboo

We investigate ways of making delay announcements in large servicesystems, such as call centers. We consider announcing the delay ofthe last customer to enter service (LES) to delayed customers, upon ar-rival. Customers typically respond to delay announcements by becom-ing more or less patient, and their response alters system performance.We explicitly model announcement-dependent customer behavior. Westudy the accuracy of the LES announcement in single-class multi-server Markovian queueing models with customer abandonment. Wesupplement our results with simulation.

3 - Speed Quality Trade-offs in a Dynamic ModelVasiliki Kostami, Management Science & Operations, LondonBusiness School, Regent’s Park, NW1 4SA, London, UnitedKingdom, vkostami@london.edu, Sampath Rajagopalan

An important trade-off organizations face is one between quality andspeed. Working faster may result in greater output and less delay butmay result in lower quality and dissatisfied customers. In this work,we consider dynamic models in a monopoly setting to explore the op-timal balance among the multiple dimensions of speed, price and waittime. The impact of quality is captured via the market demand po-tential which is a function of quality. We obtain several noteworthyresults under three scenarios: When only one feature can be changedand when both can be changed.

4 - Collaboration in Service Networks: Architecturesand ThroughputJan Van Mieghem, Kellogg School of Management,Northwestern University, 2001 Sheridan Road, Jacobs Center,5th Floor, 60208, Evanston, IL, United States,vanmieghem@kellogg.northwestern.edu

Motivated by the trend towards collaboration, we study processing net-works where some activities require the simultaneous collaboration ofmultiple human resources. Collaboration introduces resource synchro-nization requirements that are not captured in the standard procedure toidentify bottlenecks and theoretical capacity. We introduce the notionsof collaboration architecture and unavoidable idleness, due to whichthe theoretical capacity exceeds the maximal achievable throughput oractual capacity. We identify a special class of collaboration architec-tures that have no unavoidable idleness.

2

EURO-INFORMS - Rome 2013 MA-06

� MA-04Monday, 8:30-10:00O4-4

Complementarity Problems andVariational Inequalities 1

Stream: Mathematical ProgrammingInvited sessionChair: Goran Lesaja, Mathematical Sciences, Georgia SouthernUniversity, 203 Georgia Ave., 30460-8093, Statesboro, Georgia,United States, goran@georgiasouthern.edu

1 - Interior point methods for sufficient LCP in a wideneighborhood of the central pathFlorian Potra, Mathematics & Statistics, University ofMaryland, MD 21250 , Baltimore, United States,potra@umbc.eduWe present three interior-point methods for solving sufficient horizon-tal linear complementarity problems using the wide neighborhood ofAi and Zhang. The algorithms do not depend on the handicap of theproblem. They are superlinearly convergent and have optimal com-plexity. The first method is Q-quadratically convergent for problemthat have a strict complementarity solution. The second order methodis Q-superlinearly convergent with Q order 1.5 for general problems,and with Q order 3 for problems that have a strict complementaritysolution.

2 - Full NT-Step Faesible Interior-Point Method for theCartesian P(kappa)-Linear Complementarity Problemover Symmetric ConesGoran Lesaja, Mathematical Sciences, Georgia SouthernUniversity, 203 Georgia Ave., 30460-8093, Statesboro,Georgia, United States, goran@georgiasouthern.eduA feasible interior-point method (IPM) for the Cartesian P(kappa)-linear complementarity problem over symmetric cones (SCLCP) ispresented. The method uses Nesterov-Todd (NT) search directions andfull step updates of iterates. With appropriate choice of parameters thealgorithm generates a sequence of iterates in the small neighborhoodof the central path which implies global convergence of the methodand local quadratic convergence of iterates. The iteration complexityof the method matches the currently best known iteration bound forIPMs solving P(kappa)-SCLCP.

3 - Convergence of stochastic average approximationsfor a class of stochastic optimization problemsBogdan Gavrea, Faculty of Automation and ComputerScience, Department of Mathematics, Technical University ofCluj-Napoca, 15 Constantin Daicoviciu Street, 400020, Cluj -Napoca, Romania, Bogdan.GAVREA@math.utcluj.roWe present convergence results of sample average approximationsto stochastic optimization problems with mixed expectation and per-scenario constraints. The convergence results are obtained by perform-ing a sensitivity analysis for a mixed nonlinear equations-variationalinequality problem. This is a joint work with Mihai Anitescu (ANL).Acknowledgment: The work of B. Gavrea was supported by the Sec-toral Operational Program Human Resources Development 2007-2013of the Romanian Ministry of Labor, Family and Social Protection un-der the Financial Agreement POSDRU/89/1.5/S/62557.

� MA-05Monday, 8:30-10:00O4-1

Dynamic Programming I

Stream: Dynamic ProgrammingInvited sessionChair: Lidija Zadnik Stirn, Biotechnical Faculty, University ofLjubljana, Vecna pot 83, 1000, Ljubljana, Slovenia,lidija.zadnik@bf.uni-lj.si

1 - Multiobjective and dynamic programming procedurefor group decision making in resource managementLidija Zadnik Stirn, Biotechnical Faculty, University ofLjubljana, Vecna pot 83, 1000, Ljubljana, Slovenia,lidija.zadnik@bf.uni-lj.si

Resource management problems are characterized by multiple andnon-commensurable objectives over a long time horizon, by uncer-tainty and participation of different stakeholders and interest groups.Hence, issues related to compromise, negotiation and conflict are cen-tral to this dynamic decision process. This paper describes the pro-cedure in which group AHP method is integrated in a discrete dy-namic programming system. A new method for aggregating individualjudgments is proposed and a theorem about acceptable consistency isproved. A numerical example is used to illustrate the procedure.

2 - Stochastic Dual Dynamic Programming and CVaRApplied to Long Term Energy Operation PlanningAndre Marcato, Electrical Engineering Department, UFJF,CAMPUS UFJF - Faculdade de Engenharia, Sala 206 - PPEE,36036-900, Juiz de Fora, MG, Brazil,andre.marcato@ufjf.edu.br, Rafael Brandi, Reinaldo Souza,Bruno Dias, Tales Ramos, Ivo Chaves Silva Junior

This paper presents the use of CVaR (Conditional Value at Risk) to-gether with Stochastic Dual Dynamic Programming for calculation ofhydrothermal dispatch on long term horizon. The Brazilian data basecomprising all hydro-thermal system was used for to validate both theCVaR as the risk aversion curve (CAR) methodologies. The CARmethodology has been used for many years in the Brazilian energyoperation planning. It will be shown that using the CVaR is more ap-propriate for this problem getting operational scenarios safer.

3 - A Novel Dynamic Programming Approach for Solvingthe Container Pre-Marshaling ProblemMatthias Prandtstetter, Mobility Department, DynamicTransportation Systems, AIT Austrian Institute of TechnologyGmbH, Giefinggasse 2, 1210, Vienna, Austria,matthias.prandtstetter@ait.ac.at

The Pre-Marshaling Problem (PMP) arises in container terminalswhere the goal is to find an operation plan with as few container move-ments as possible for pre-ordering containers according to specifiedpriorities such that containers with higher priorities are not blocked bycontainers with lower ones. We propose a new dynamic programmingapproach for solving the PMP. Computational experiments emphasizethat this is the first exact approach being capable of solving (almost) allbenchmark instances of real-world size available in literature to provenoptimality within only a few milliseconds.

� MA-06Monday, 8:30-10:00O4-2

Recent Advances in Global Optimization 1

Stream: Global OptimizationInvited sessionChair: Herman Mawengkang, Mathematics, The University ofSumatera Utara, FMIPA USU, KAMPUS USU, 20155, Medan,Indonesia, mawengkang@usu.ac.idChair: Elise del Rosario, OSSFFI, 14A CyberOne Bldg.,Eastwood, Libis, 1110, Quezon City, Metro Manila, Philippines,elise.del.rosario@stepforward.phChair: Gerhard-Wilhelm Weber, Institute of Applied Mathematics,Middle East Technical University, ODTÜ, 06531, Ankara, Turkey,gweber@metu.edu.tr

1 - Direct search approach for solving non-convexmixed-integer non linear programming problemsHardi Tambunan, Mathematics, Quality University, FMIPAUSU, 20155, Medan, North Sumatera Province, Indonesia,harditambunan29@yahoo.com, Herman Mawengkang

3

MA-07 EURO-INFORMS - Rome 2013

The mixed integer nonlinear programming problem addressed in thispaper has a structure characterized by a subset of variables restricted toassume discrete values, which are linear and separable from the contin-uous variables. The basic idea is to use a strategy of releasing nonbasicvariables from their bounds found in the optimal continuous solution insuch a way to force the appropriate non-integer basic variables to moveto their neighbourhood integer points. Some computational experienceare presented.

2 - Stochastic programming model for Land use man-agement problemsSiti Rusdiana, Mathematics, University of Syiah Kuala,FMIPA USU, 20155, Medan, North Sumatera Province,Indonesia, srusdiana10@yahoo.com

Land is used to meet a multiplicity and variety of human needs andto serve numerous, diverse purposes. Our study is focused findingan optimal land/resources portfolio composition through time, in thepresence of future market uncertainty. In this paper we formulatea scenario-based stochastic integer programming model, which takesinto account the uncertainty related to the market value of revenues ac-cruing from the land in different states In order to take into account thenon-constant incremental benefits accruing from different land alloca-tions.

3 - Sampling and integer programming approach forsolving water distribution network with reliabilityconsiderationAsrin Lubis, Mathematics, Unimed/Graduate School ofMathematics, University of Sumatera Utara, FMIPA USU,20155, Medan, Sumatra Utara, Indonesia,lubis.asrin@yahoo.com

We propose a nonlinear stochastic optimization model for tackling aWDN under the consideration of reliability in water flows. The non-linearities arise through pressure drop equation. We adopt samplingand integer programming based approch for solving the model. A di-rect search algorithm is used to solve the integer part.

4 - Scrambling index of a class of 2-digraphsMulyono Mulyono, Mathematics, University NegeriMedan/Grad. School of Math. USU, Fmipa usu, 20155,Medan, Indonesia, mulyonomulyono162@yahoo.com, SaibSuwilo

A 2-digraph D is a digraph in which each of its arcs is colored by red orblue. The srambling index of 2-digraph D, denoted k(D), is the small-est positive integer s+t over all nonnegative integers s and t such thatfor each pair of vertices u and v in D there is a vertex w with the prop-erty that there is walks from u to w and from v to w consisting of s redarcs and t blue arcs. We discuss the scrambling index for 2-digraphconsisting of two cycles whose lengths differ by 1.

� MA-07Monday, 8:30-10:00O4-3

Vector and Set-Valued Optimization I

Stream: Vector and Set-Valued OptimizationInvited sessionChair: César Gutiérrez, Departamento de Matemática Aplicada,Universidad de Valladolid, E.T.S.I. Informática, Co. Cementerios/n, Campus Miguel Delibes, 47011, Valladolid, Spain,cesargv@mat.uva.esChair: Vicente Novo, Matematica Aplicada, Universidad Nacionalde Educacion a Distancia, Juan del Rosal no. 12, 28040, Madrid,Spain, vnovo@ind.uned.es

1 - Generating epsilon-optimal solutions in multiobjec-tive optimization by genetic algorithmMarcin Studniarski, Faculty of Mathematics and ComputerScience, University of Lodz, ul. S. Banacha 22, 90-238, Lodz,Poland, marstud@math.uni.lodz.pl, El-Desouky Rahmo

In [1], some stopping criterion for a genetic algorithm was established,which enables one to find, with a prescribed probability, all minimalsolutions in a finite multiobjective optimization problem. In this pa-per, we show how to apply the above-mentioned result to generateepsilon-optimal solutions to some continuous multiobjective optimiza-tion problem.

[1] M. Studniarski, Finding all minimal elements of a finite partiallyordered set by genetic algorithm with a prescribed probability, Numer-ical Algebra, Control and Optimization 1 (2011), 389-398.

2 - Coercivity, stability, and well-posedness in multiob-jective optimizationRuben Lopez, Departamento de Matematica y FisicaAplicadas (DMFA), Universidad Catolica de la SantisimaConcepcion, Alonso Ribera 2850, 409-0541, Concepcion,VIII Region, Chile, rlopez@ucsc.clIn this talk we show how the notions of coercivity, stability, and well-posedness are related in multiobjective optimization theory. To do this,we employ a recent notion of variational convergence for vector-valuedfunctions.

* This work has been supported by Proyecto FONDECYT 1100919through CONICYT-Chile.

3 - Approximate proper solutions of vector optimizationproblems. Properties and limit behaviourLidia Huerga, Applied Mathematics, UNED, C/Juan delRosal, 12, 28040, Madrid, Spain, lhuerga@bec.uned.es, CésarGutiérrez, Bienvenido Jiménez, Vicente NovoWe present a new notion of approximate proper solution for a con-strained vector optimization problem on a locally convex topologicalvector space. We study the properties of this type of approximateproper solutions when the error is fixed and also when it tends to zero.In particular, we show that these solutions are useful in order to obtainouter approximations of the efficient set without considering convexityassumptions.

4 - A new concept of solution in vector optimization de-fined via improvement setsCésar Gutiérrez, Departamento de Matemática Aplicada,Universidad de Valladolid, E.T.S.I. Informática, Co.Cementerio s/n, Campus Miguel Delibes, 47011, Valladolid,Spain, cesargv@mat.uva.es, Bienvenido Jiménez, VicenteNovoIn this talk we introduce a new notion of solution for vector optimiza-tion problems based on improvement sets. This notion generalizes themost usual concepts of non-dominated solution for this kind of opti-mization problems. We characterize these solutions by scalarizationin convex and non-convex problems. In the first case, the results areobtained by assuming generalized convexity assumptions, and in thesecond one, they are consequence of characterizing the improvementsets as sublevel sets of nonlinear mappings.

� MA-11Monday, 8:30-10:00G5-3

System Dynamics Modelling andSimulation Session 1Stream: System Dynamics Modeling and SimulationInvited sessionChair: Rogelio Oliva, Mays Business School, 301C Wehner - 4217TAMU, 77843-4217, College Station, Texas, United States,roliva@tamu.edu

1 - Modeling bounded rationality: analysis of the con-sumer choice in the adoption of alternative fuel vehi-clesLaura Ardila, Computer and Decision Science Department,Universidad Nacional de Colombia, Carrera 80 No 65-223,Medellín, Antioquia, Colombia, laardilaf@unal.edu.co,Carlos Jaime Franco

4

EURO-INFORMS - Rome 2013 MA-13

This paper presents an approximation to the modeling of bounded ra-tionality (BR) in the adoption of alternative-fuel vehicles in the Colom-bian Market. In order to do this, a simulation model was built. Fromthe results is possible analyze the importance of including BR in themodeling process. The formulated model using BR explains ade-quately the behavior within the system; the model allows a better com-prehension of the consumer decisions, and helps with strategy’s con-struction. The model can be adapted for other economies with similarcharacteristics to the Colombian market.

2 - Analysis of the mechanism Reliability Charge in theColombian electricity market: a system dynamic ap-proachEstefany Garcés Arango, Ciencias de la computación y ladecisión, Universidad Nacional de Colombia, Colombia,egarces@unal.edu.co, Carlos Jaime Franco, Isaac DynerRezonzew

One of the main problems facing liberalized electricity markets is to es-tablish how the generators must be paid to guarantee the investments innew capacity and to ensure the system reliability. Therefore, the mar-ket regulator has decided to implement a security mechanism calledReliability Charge. The results of this new regulation are not obvious,because Colombia has a high dependence on hydro-electric plants. Inthis paper, we explain a System Dynamic model build for analyzingthe effects involving the implementation of this Reliability Charge inthe Colombian electricity market.

3 - Three strategic views of organizations. Implicationsfor their dynamics.Bent Erik Bakken, Norwegian Defence University College,St. Georgs vei 4, 0280 OSLO, 0280 OSLO, Norway, Oslo,Norway, beerikba@online.no

A common strategic view of organizations as value chains has beenchallenged by two complementary value configuration views; valueshops and value networks. This paper discusses where each perspec-tive adds value, and the advantages and disadvantages of allowing 1, 2or 3 configurations to be used in parallel. One system dynamics modelportraying each of the three perspectives are provided for a defence or-ganization. Reference modes and sensitivity analysis are provided foreach value configuration, and implications for organizational designdiscussed.

4 - Structural dominance in large and stochastic systemdynamics modelsRogelio Oliva, Mays Business School, 301C Wehner - 4217TAMU, 77843-4217, College Station, Texas, United States,roliva@tamu.edu

Duggan and Oliva (2013) report a significant effort to develop methodsfor identifying structural dominance in system dynamics models. Todate, however, the testing of these methods has been in small (less than10 stocks) deterministic models that show smooth transitions. Whilethe analysis of simple models is necessary step for proof of concept, themethods have become stable enough to be tested in a wider domain. Inthis paper we report the findings from expanding the application thesemethods in two dimensions: increasing model size and incorporatingstochastic variance in model variables.

� MA-12Monday, 8:30-10:00G5-4

Facility location

Stream: Transportation and LogisticsInvited sessionChair: Bernard Gendron, DIRO/CIRRELT, Université deMontréal, C.P. 6128, succ. Centre-ville, H3C 3J7, Montréal,Québec, Canada, gendron@iro.umontreal.ca

1 - Dynamic Facility Location with Generalized ModularCapacities

Sanjay Dominik Jena, CIRRELT, Université de Montréal,Canada, sanjay.jena@cirrelt.ca, Bernard Gendron,Jean-François Cordeau

We introduce a facility location problem with multiple time periodsand multiple capacity levels, where the costs for capacity changes arebased on a cost matrix. A strong mixed integer programming model ispresented that unifies several existing problems found in the literature.We apply our model to three special cases. Computational experimentsshow that our model can obtain optimal solutions in shorter comput-ing times than the existing specialized formulations. Furthermore, wepresent a Lagrangean Relaxation heuristic to find good quality solu-tions in even shorter computation times.

2 - A primal heuristic and a Lagrangian relaxation for anindustrial two-echelon location-distribution problemPaul-Virak Khuong, DIRO/CIRRELT, Université de Montréal,H3C 3J7, Montreal, Canada, pvk@pvk.ca, Bernard Gendron

We describe practical methods to solve an industrial location-distribution problem approximately, with solution quality estimates.The problem pertains to the operation of an adaptive multi-echelondistribution system: intermediate locations consolidate flows, and areadjusted in response to changes in demand. A MIP formulation cap-tures the problem’s complex costs, but is very large. We present amultilayer variable neighbourhood search and a Lagrangian decompo-sition method for the same model: the former quickly yields feasiblesolutions, while the latter proves their quality.

3 - Lagrangian Relaxation Approaches for Multicom-modity Uncapacitated Network DesignEnrico Gorgone, DIMES - Dipartimento di IngegneriaInformatica Modellistica Elettronica e Sistemistica, Universitàdella Calabria, Via Bucci, CUBO 41c, VI piano, 87036,Rende, Cosenza, Italy, egorgone@deis.unical.it, AntonioFrangioni, Bernard Gendron

We present several algorithms for solving the multicommodity un-capacitated network design problem. We report the results obtainedby using different Lagrangian relaxation approaches, a classical spe-cialized dual-ascent method and combinations of these algorithms.We solve the Lagrangian duals by using different variants of nondifferentiable optimization approaches like (incremental, deflected,projected) subgradient-type methods and (disaggregated, generalized)bundle type methods.

4 - The Impact of Filtering in a Branch-and-Cut Algo-rithm for Multicommodity Capacitated Network De-signBernard Gendron, DIRO/CIRRELT, Université de Montréal,C.P. 6128, succ. Centre-ville, H3C 3J7, Montréal, Québec,Canada, gendron@iro.umontreal.ca, Mervat Chouman,Teodor Gabriel Crainic

We study the impact of different filtering methods embedded into aspecialized branch-and-cut algorithm for the multicommodity capac-itated network design problem. Contrary to the preprocessing tech-niques used in state-of-the-art MIP solvers, these filtering methods ex-ploit the structure of the problem, while being applicable to a verylarge class of network design problems. Three types of filtring meth-ods are developed: preprocessing and postprocessing at each node ofthe tree, as well as domination rules among different nodes of the tree.Computational results will be presented.

� MA-13Monday, 8:30-10:00G5-5

Facility Logistics and Order-Picking in aWarehouseStream: Facility LogisticsInvited sessionChair: Pierre Baptiste, de mathématiques et de Génie Industriel,École Polytechnique de Montréal, CP. 6079, succ. centre-ville„H3C 3A7, Montréal, Québec, Canada, pbaptiste@polymtl.ca

5

MA-14 EURO-INFORMS - Rome 2013

1 - Ergo-picking: multi-objective approach to considerergonomics aspects in picking systems design andmanagementFabio Sgarbossa, Department of Management andEngineering (DTG), Univerity of Padova, 36100, Vicenza,Italy, fabio.sgarbossa@unipd.it, Daria Battini, AlessandroPersonaOne of the most time consuming activities in the logistics, with im-portant energy expenditure, is picking. Traditional optimization ap-proaches considers only time variables. In this paper an innovativemulti-objective model has been developed to design picking systemsconsidering also the ergonomics aspect, defined by energy expendi-ture, based on main features of these systems, such as: location, char-acteristics of the piece and handling systems. Several real case studieshave allowed the validation using an innovative motion capture systemwith an integrated ergonomic evaluation tool.

2 - A multi period order picking approach to a multistage logistic systemRifat Gürcan Özdemir, Industrial Engineering Department,Istanbul Kültür University, Atakoy Campus,Atakoy-Bakirkoy, 34156, Istanbul, Turkey,rg.ozdemir@iku.edu.tr, Ezgi Nilay Uzun, Nihan TopukIn this paper, a multi period order picking approach to a multi stage lo-gistic system is developed and aimed to determine optimal order pick-ing strategy. The logistic system involves three components as follows:A plant that sends items to the warehouse in which orders are pickedand shipped to the customers. The most critical issue for the entiresystem is planning order picking process in the warehouse efficiently.For this purpose, a mathematical model is developed for minimizingnumber of pickers subject to lumpy orders received from customers,and dynamic shelf capacities.

3 - Integrated Batching and Routing in Warehouses withOrders and ReturnsSusanne Wruck, Distribution and Supply Chain Logistics, VUUniversity Amsterdam, De Boelelaan 1105, 1081 HV,Amsterdam, Netherlands, s.wruck@vu.nlWe present an iterated local search approach for the simultaneous op-timization of order batching and order picker routing in warehouses.Thereby we incorporate deadlines of customer orders and schedulea working shift for multiple order pickers simultaneously. Further-more, we incorporate the restoring of potential product returns duringthe picking tours. We demonstrate that significant cost savings canbe achieved by the integration of order and return processing and thatjoint batch and route optimization can lead to better order picking per-formance.

4 - Multiple doors cross docks schedulingPierre Baptiste, de mathématiques et de Génie Industriel,École Polytechnique de Montréal, CP. 6079, succ.centre-ville„ H3C 3A7, Montréal, Québec, Canada,pbaptiste@polymtl.ca, Mohammad yousef MaknoonMaterial handling inside cross dock plays a critical role. To haveefficient handling plan, transferring decisions should be simultane-ously considered with the trucks processing order, incoming and out-going.We present a mathematical model of the material handling ina cross dock with multiple doors (optimal solution up to 6 doors, 20trucks with Cplex) and an heuristic for bigger size problems. Numer-ical results depict the stability of this heuristic and compare both ap-proaches. Finally, we investigate various loading scenarios and providesome recommendations on material handling decisions.

� MA-14Monday, 8:30-10:00G5-6

MetaheuristicsStream: MetaheuristicsInvited sessionChair: Alok Choudhary, Management School, University ofSheffield, IWP Building, Mushroom Lane, S10 2TN, Sheffield,United Kingdom, a.choudhary@sheffield.ac.uk

1 - GRASP and Path Relinking based on structured so-lution clusteringArne Løkketangen, Molde University College, Britveien 2,6411, Molde, Norway, arne.lokketangen@himolde.no,Jianyong Jin

Solution clustering based on structural similarity is a novel way to par-tition elite solutions found during a search. The structural informationcaptured provides superior guidance for later intensification phases.We explore different structurally based solution distance measures, aswell as novel variants of GRASP and Path Relinking based on utiliz-ing information from structured solution clustering. Both in-clusterand between-cluster guidance is investigated. The methods are testedon large benchmark VRPs from the literature.

2 - Optimization of microchannels for liquid cooling ofelectronic chipsTuba Okutucu, Mechanical Engineering, Middle EastTechnical University, Makina Mühendisliği Bölümü, A-143,ODTU, 06800, Ankara, Turkey, okutucu@metu.edu.tr, GÖkerTÜrkakar

Dimensional optimization of silicon microchannel heat sinks is per-formed by minimizing the total thermal resistance. Intel Core i7-900Desktop Processor is considered as the reference chip to be cooled.The study is unique in that the optimization has been performed forlocalized multiple heat sources. The results of the optimization agreedvery well with available ones in the literature. Grid search is utilized toobtain the minimum value of the objective function. The optimizationprocedure may be improved using a meta-heuristic algorithm such asTabu search.

3 - Progressive multi-objective optimizationKenneth Sörensen, Faculty of Applied Economics, Universityof Antwerp, Prinsstraat 13, 2000, Antwerpen, Belgium,kenneth.sorensen@ua.ac.be, Johan Springael

PMOO is a novel technique that includes the decision maker’s prefer-ences into the multi-objective optimization process instead of tacklingthese steps sequentially. In this talk, we present a PMOO method forthe multi-objective knapsack problem. This approach integrates thewell-known PROMETHEE multi-criteria method into a simple tabusearch method.

4 - A carbon-footprint optimization model for integratedforward-reverse logisticsAlok Choudhary, Management School, University ofSheffield, IWP Building, Mushroom Lane, S10 2TN,Sheffield, United Kingdom, a.choudhary@sheffield.ac.uk

This research proposed a MILP optimization model for forward-reverse logistics with carbon-footprint considerations. This researchextends the traditional model by both minimizing cost and carbon foot-print across the supply chain and incorporates carbon emission param-eters with various operational decision-making variables related to fa-cility layout decisions, procurement and production. We provide a se-ries of insights that highlight the impact of operational decisions oncarbon emissions. An efficient forest data structure algorithm is usedto solve the optimization problem.

� MA-15Monday, 8:30-10:00G5-2

Genetic Algorithms and GeneticProgramming

Stream: Metaheuristics (contributed)Contributed sessionChair: Derya Dengi, Department Industrial Engineering, YasarUniversity, Yasar University Selcuk Yasar Kampusu No:35-37,Agacliyol Bornova, 35100, Izmir, Izmir, Turkey,deryadengi@gmail.com

6

EURO-INFORMS - Rome 2013 MA-17

1 - Comparative Analysis of Differential Evolution, Fire-fly and Genetic Algorithms on Dynamic Environ-mentsBurcin Ozsoydan, Industrial Engineering, Dokuz EylulUniversity, Turkey, burcin.ozsoydan@deu.edu.tr, AdilBaykasoğlu

Most of the real-life problems require a dynamic optimization aris-ing from the uncertainty of future events, which makes optimizationharder. Genetic Algorithms (GA) and Differential Evolution (DE) havebeen widely used for static optimization. Their applications to dynamicoptimization problems are rare. In the present work an improved Fire-fly Algorithm (FA) is presented for dynamic optimization along withcomparisons via GA and DE. Results demonstrate that, the proposedFA significantly achieves superior results for both stationary and dy-namic problems in comparison to GA and DE.

2 - Optimizing intrusion detection system performancein ubiquitous environments using genetic algorithmapproachLynda Sellami, computer science, University of Bejaia, routede targua ouzemou, 06000, Bejaia, Algeria, slynda1@yahoo.fr

The objective of ubiquitous computing is to give the users the abilityto communicate and provide information regardless of their locationin the Internet network, The success of these distributed computingdepends largely on a strong security mechanism such as intrusion de-tection. This article presents a review to understand, examine, and cri-tique the current state of its applications in various fields and proposinga solution to this problem by using genetic algorithm approach.

3 - Internet decision system for combinatorial optimiza-tion problems under permutationLamia Trabelsi, Laboratoire ROI Ecole Polytechnique,Tunisia, lamia_tr2001@yahoo.fr, Talel Ladhari

In this work, we propose a web-based decision support system to solvecombinatorial optimization problems under permutation (COPP) usinggenetic programming. The REST (Representational State Transfer) ar-chitecture style is used to allow the new system to communicate withresearcher applications over the net. REST avoids inter-operabilityproblem. It allows accessing resources by URI (Universal ResourceIdentifier). Via the URI, researchers are able to reuse or to evaluate atrun-time their operators with the available operators.

4 - A Genetic Algorithm Based Approach for Opera-tor Allocation Problem in Garment Industry with theConsideration of Labour Skill LevelsDerya Dengi, Department Industrial Engineering, YasarUniversity, Yasar University Selcuk Yasar KampusuNo:35-37, Agacliyol Bornova, 35100, Izmir, Izmir, Turkey,deryadengi@gmail.com, Gonca Tuncel

Sewing lines in garment industry are complicated assembly lines dueto the labour intensive work nature with great number of operations,various types of machines, and different labour skill levels. In thisstudy, we considered operator allocation problem in an internationalgarment factory. Labour skill levels in each of the operation are alsotaken into account to improve the workload balance. The problem issolved by using a genetic algorithm based approach. The experimentalresults for different product models are presented and discussed.

� MA-16Monday, 8:30-10:00G5-7

Network Flows IStream: Routing ProblemsInvited sessionChair: Paola Festa, Dept. of Mathematics and Applications,University of Napoli Federico II, Compl. MSA - Via Cintia,80126, Napoli, Italy, paola.festa@unina.itChair: Francesca Guerriero, D.E.I.S.: Department of Electronics,Computer Science and Systems, University of Calabria, Via P.Bucci, 87036, Rende, Italy, guerriero@deis.unical.it

1 - Solving the Shortest Path Tour Problem and Some ofIts VariantsPaola Festa, Dept. of Mathematics and Applications,University of Napoli Federico II, Compl. MSA - Via Cintia,80126, Napoli, Italy, paola.festa@unina.it, Daniele Ferone,Francesca Guerriero, Demetrio Laganà

The shortest path tour problem consists of finding in a weighted di-graph a shortest path from a given origin node to a given destinationnode, crossing a sequence of disjoint and possibly different-sized nodesubsets, given in a fixed order. In this talk, a polynomial-time reduc-tion of the problem to a classical shortest path problem over a modifieddigraph is illustrated. Several exact solution methods are proposed andtheir performances are compared with the state-of the-art solving pro-cedure. Some variants of the problem and their properties are analyzed.

2 - Maximizing the propagation of the advertisements insocial networksCiriaco D’Ambrosio, Department of Computer Science,University of Salerno, Italy, cirodamb@gmail.com, FrancescoCarrabs, Raffaele Cerulli

We study the problem of maximizing the propagation of an advertise-ment in a social network where a node represents a user and an edgerepresents the friendship relation. The problem is to identifying a setof nodes in which to publish an advertisement in order to maximizeits propagation without wasting ad spaces. Moreover, we analyze an-other problem where the aim is to maximize the number of selectednodes ensuring a fixed propagation level. Two formulations and ametaheuristic approach are presented.

3 - Fast Shortest Path Queries in Road NetworksLeonid Antsfeld, NICTA, 223 Anzac Pde., NICTA, Level 4,Reception, 2052, Kensington, NSW, Australia,leonid.antsfeld@nicta.com.au, Philip Kilby, Andrew Verden,Toby Walsh

Our new algorithm, identifies key nodes in a road network. Initially,we identify all major nodes, i.e., nodes which have at least one end-point as a highway. Next, we divide our map to clusters. Then forevery cluster, we identify the furthest major node, such that there is noother major node on a shortest path from this cluster. The set of accessnodes of a cluster is defined as a hitting set of all major nodes whichare on a shortest paths originating from the convex hull of the cluster.On average a cluster has 5 access nodes, therefore the query translatesto just a few table lookups.

4 - Network interdiction by dynamic interdictor usingshortest pathAhmet Kabarcik, Industrial Engineering, Cankaya University,Eskisehir Yolu 29.km, Yenimahalle, 06810, Ankara, Turkey,a.kabarcik@cankaya.edu.tr, Levent Kandiller, Haluk Aygunes

This study considers a network interdiction problem in which arcs aredisabled by a dynamic interdictor. Its assumed that the network is com-posed of two interconnected networks. One of the networks is used bythe user and the other one is used by the interdictor. An interdictor’snode may either be deployed on an arc or on a node of the networkuser. Nodes on the interdictor’s network are candidate points for de-struction. Here, it’s aimed to find the shortest path that can be used bythe interdictor while destroying all paths between origin and destina-tion nodes of network user.

� MA-17Monday, 8:30-10:00G5-8

Green Vehicle Routing

Stream: Sustainable Transport PlanningInvited sessionChair: Esben Linde, DTU Transport, Technical University ofDenmark, Bygning 115, Bygningstorvet, 2800, Kgs. Lyngby,Denmark, esli@transport.dtu.dk

7

MA-19 EURO-INFORMS - Rome 2013

1 - Environmental Vehicle Routing Problem (VRP)Georgios K.D. Saharidis, Mechanical Engineering, Universityof Thessaly, Pedion Areos, 38334, Volos, Thessaly, Greece,saharidis@gmail.com, George Liberopoulos

The vehicle is a very important component for the evaluation and es-timation of environmental externalities in the transport sector. How-ever, there is another important component that affects the amount ofemissions that has been taken either partially or not at all under consid-eration. This component is related to the characteristics of the trans-portation network (TN). In this research, we develop a novel objectivefunction for the environmental VRP that takes as input several data andestimates the Environmental Externalities Score (EES) of a route.

2 - Demand Management and Vehicle Routing in E-fulfilmentRichard Eglese, The Management School, LancasterUniversity, Department of Management Science, LA1 4YX,Lancaster, Lancashire, United Kingdom,R.Eglese@lancaster.ac.uk, Xinan Yang, Arne Karsten Strauss,Christine Currie

For many customers, ordering goods over the internet for home deliv-ery offers a more sustainable way of shopping than making a specialtrip by car to a retail outlet. However attended home delivery servicesface the challenge of providing narrow delivery time slots to ensurecustomer satisfaction, whilst keeping the delivery cost under control.Dynamic pricing policies based on a customer choice model are pro-posed to determine how much incentive (discount or charge) to offerfor which time slot. A simulation study based on real data shows theeffectiveness of the approach.

3 - An Algorithm for Sustainable Vehicle RoutingFernando Oliveira, Operations MAnagement, ESSECBusiness School, Cergy, France, oliveira@essec.fr, AnneLiret

The vehicle routing problem is known for being NP-hard. In this ar-ticle we analyze this problem from a perspective of a company thatwants to introduce vehicle sharing in an attempt to increase efficiencyand reduce CO2 emissions. We provide a formulation of the problemconsidering multiple depots and time windows. We propose an algo-rithm to approximate the solution is a real-case study and analyze itsmain properties.

4 - Routing of Electric Vehicles: City Distribution inCopenhagenEsben Linde, DTU Transport, Technical University ofDenmark, Bygning 115, Bygningstorvet, 2800, Kgs. Lyngby,Denmark, esli@transport.dtu.dk, Allan Larsen, AndersVedsted Noerrelund, Stefan Ropke, Min Wen

In this work, a Vehicle Routing Problem with Time Windows consid-ering EV constraints of limited driving range and freight capacity isaddressed (EVRPTW). The EVs are allowed to recharge at certain lo-cations, and aspects of intelligent location of these recharging pointsare considered. The objective is to find the least cost plan for EV rout-ing and compare this to conventional routing. A heuristic method isdeveloped and tested on data based on real-life collected data on distri-bution vehicles in central Copenhagen, Denmark. The EVRPTW hasso far received little attention in the literature.

� MA-19Monday, 8:30-10:00G5-10

Advanced Inventory Control in SCM

Stream: Business Excellence in LogisticsInvited sessionChair: Sebastian Langton, Logistics Management Department,Helmut-Schmidt-University, 22041, Hamburg, Germany,langton@hsu-hh.deChair: Johannes Siebert, Business Administration, University ofBayreuth, Universitätsstr. 30, 95440, Bayreuth, Bayern, Germany,Johannes.Siebert@uni-bayreuth.de

1 - Inventory optimization for a recoverable manufactur-ing system with product substitutionSemra Sebnem Ahiska, Industrial Engineering Department,Galatasaray University, Ciragan Cad. No:36 Ortakoy, 34349,Istanbul, Turkey, sahiska@gsu.edu.tr, Emre Kurtul, RustyKing

The periodic-review inventory control problem for a stochastic manu-facturing/remanufacturing system with product substitution is studied.A remanufactured item has a lower selling price than a manufactured(i.e., new) item, thus separate demand streams exist for those products.A product substitution strategy is considered where the demand for aremanufactured item is satisfied by a new item if no remanufactureditem is left in stock and new item inventory is available. Through aMarkov decision analysis, we find the optimal inventory policies, anddetermine the value of product substitution.

2 - Decision Making under Bounded Rationality Terms:Application in a Transportation ProblemMaria Giannopoulou, Industrial Management & Technology,University of Piraeus, Department of Industrial Management& Technology University of Piraeus — 18534 Piraeus,GREECE, 18534, Piraeus, Greece, mgiannopulu@gmail.com,Dimitrios Emiris

Decision makers operate under cognitive, emotional and computa-tional constraints in unstable and complicated environments. TheBounded Rationality (BR) theory takes into account such constraintsand weaknesses to provide a more realistic decision making frame-work. This work presents a Fuzzy Cognitive Map (FCM) for a simu-lation problem of strategy optimization in transportation contracts soldby e-auctions. The model inputs are concepts concerning BR elementsaffecting bidders’ behavior and generates data concerning the interac-tions between these concepts and potential correlations among them.

3 - A Two-Loop Machine Learning Approach for Evaluat-ing Supply Chain Stockout EventsSebastian Langton, Logistics Management Department,Helmut-Schmidt-University, 22041, Hamburg, Germany,langton@hsu-hh.de, Johannes Siebert, Martin Josef Geiger

Despite the relevance of stockout cost parameters in many logisticsmodels, a need for practicable cost quantification methods can stillbe stated. As a methodologically new idea we present an interac-tive machine learning approach for estimating such costs. Previousexperiments revealed that the consistency of data from the decisionmaker (DM) is essential for satisfying machine learning results. Inline with this, we introduce an idea of inconsistent data elimination byre-integrating the DM in a second learning loop. The talk completeswith presenting experimental work and numerical results.

� MA-20Monday, 8:30-10:00G5-11

Routing in Public Transport Networks

Stream: Optimization in Public TransportInvited sessionChair: Gianpaolo Oriolo, Universita’ Roma Tor Vergata, Italy,oriolo@disp.uniroma2.itChair: Giuseppe Italiano, Department of Civil Engineering andComputer Science Engineering, University of Rome "TorVergata", via del Politecnico 1, 00133, Rome, Italy,italiano@disp.uniroma2.it

1 - Stochastic Route Planning in Public TransportAlpár Jüttner, IBM-ELTE Center for Applied Mathematics,Dept of Operations Research, ELTE, Pázmány P. sétány, 1/C,H-1117, Budapest, Hungary, alpar@cs.elte.hu, Kristof Berczi,Mátyás Korom, Jacint Szabo, Tim Nonner, Marco Laumanns

There are plenty of public transport planners that help passangers tofind a route satisfying their needs. However, a common weakness

8

EURO-INFORMS - Rome 2013 MA-22

of these methods that they assume a deterministic environment. Ve-hicles in public transport are typically behind or before time, hencemaking the availability of a given journey in real life questionable.We present an algorithm that overcomes this uncertainty by using astochastic model for departure and travel times. The output of the al-gorithm is not a single route but a policy for each node that defineswhich services to take at a given time.

2 - Stochastic on-time arrival problem on public trans-portation networksSebastien Blandin, IBM Research Collaboratory – Singapore,IBM Singapore Pte Ltd, 9 Changi Business Park Central, TheIBM Place, 486048, Singapore, Singapore, Singapore,sblandin@sg.ibm.com, Samitha Samaranayake, Alex Bayen

We formulate the stochastic on-time arrival problem on public trans-portation networks. Link travel times are estimated according to aroute-based headway diffusion model, which allows the computationof associated waiting times at connection nodes. We prove that this for-mulation leads to efficient computations tractable for large networks,even in the case of time-varying headway distributions. Additionally,robustness to recurrent variability as well as unexpected events is verynaturally represented under this formulation. Algorithm performanceis illustrated using realistic transit data.

3 - When Algorithms Come True: A Dynamic MultimodalRoute Planner for RomeLuca Allulli, Centrale della Mobilità, Roma servizi per lamobilità srl, Piazzale degli Archivi, 40, 00144, Rome, Italy,luca.allulli@agenziamobilita.roma.it

"Roma servizi per la mobilità" is the municipal agency for public andprivate transport in Rome. The agency recently developed a dynamicmultimodal route planner (http://muovi.roma.it). Real-time data is col-lected from GPS bus trackers, which cover the vast majority of busesin Rome. We present the main challenges our small development teamhad to face: creating a simple, yet flexible framework for dynamicmultimodal routing, which allows us to rapidly prototype and integratediverse means of transport; dealing with raw data from bus trackers;and tuning code to obtain adequate performance.

4 - Real-Time Public Transportation Route PlannersFederico Santaroni, Department of Civil Engineering andComputer Science Engineering, Tor Vergata, Via Politecnico1, 00133 , Roma, Italy, federico.santaroni@gmail.com, LuigiLaura, Federico Cosentino, Emanuele Paracone

In the last years we have seen a huge contribution to routing algorithmsin large transportation networks, especially focusing on road networks.Despite many results in this direction, if we switch our attention topublic transportation networks, this is a completely different scenarioin which one cannot directly apply the same algorithmic techniques.In this talk we will discuss the challenges arising in designing a pub-lic transportation route-planner, in particular in the case where one canuse data provided by the GPS vehicles’ devices in real time.

� MA-21Monday, 8:30-10:00G6-1

Scheduling Approaches for ComplexManufacturing and Service Systems

Stream: SchedulingInvited sessionChair: Lars Moench, FernUniversität in Hagen, 58097, Hagen,Germany, lars.moench@FernUni-Hagen.de

1 - A Reinforcement Learning Approach to Solving Hy-brid Flexible Flowline Scheduling ProblemsDmitry Borodin, Optimization, OM Partners, Koralenhoeve23, 9000, Wommelgem (Antwerp), Belgium,dk.borodin@gmail.com, Bert Van Vreckem, Ann Nowe, KrisDockx, Wim De Bruyn

This research presents a technique based on Learning Automata (LA)to solve Hybrid Flexible Flowline Scheduling Problems with addi-tional constraints like sequence dependent setup times, precedence re-lations between jobs and machine eligibility that occur in complex real-life problems like those in process industry. In the proposed method,LA play a dispersion game to determine the order of jobs to be pro-cessed so that makespan is minimized while avoiding illegal permuta-tions. Computational results on benchmark problems indicate that LAcan yield better results than the ones known until now.

2 - Missing operations in reentrant flow-shop problemsRichard Hinze, TU Dresden, Germany,richard.hinze@mailbox.tu-dresden.de, Dirk Sackmann, UdoBuscher, Gerhard Aust

We propose a model for the reentrant flow-shop scheduling in order tominimize makespan. Reentrant flow shop scheduling problems occurin practical applications, such as the manufacturing of semiconduc-tors, airplane engines, and petrochemical production. The formulationconsiders missing operations. Compared to traditional formulationsthe values of makespan are significantly reduced, due to an appropri-ate dealing with processing times equal to zero. Computational testswith different problem sizes reveal that computation times are mainlyinfluenced by the total number of jobs.

3 - A new method for the evaluation of a neighbor so-lution in local search procedures for solving theJSPTWTJens Kuhpfahl, School of Economics and Business, Chair ofProduction and Logistics, MLU Halle-Wittenberg, Germany,jens.kuhpfahl@wiwi.uni-halle.de, Christian Bierwirth

Efficient local search in job shop scheduling requires estimating thesuccess of possible neighborhood moves. Typically, a lower bound iscalculated for a new schedule by estimating heads and tails of the op-erations affected by the neighborhood search operator. In this talk weshow how this technique can be improved for the JSPTWT regardingreliability and computation time. It allows us to determine the true ob-jective function value of the neighbor solution instead of just verifyinga necessary condition for descent moves in the local search.

4 - An Instance Generator for Scheduling Problems withAlternative Process PlansOleh Sobeyko, University of Hagen, 58097, Hagen, Germany,Oleh.Sobeyko@FernUni-Hagen.de, Roman Capek, ZdenekHanzalek, Lars Moench

We consider scheduling problems with alternative process plans. Anew generator for problem instances is proposed. The generator isdesigned for both resource constrained project scheduling problemsand problems based on the integrated process planning and schedul-ing model while the data representation for both classes of problemsis unified. We propose a common model for scheduling problems withalternatives and fill the gap in existing tools for generating probleminstances. Structural properties, the resource environment, and the at-tributes of activities are defined as input parameters.

� MA-22Monday, 8:30-10:00G6-2

Scheduling & Project Management

Stream: Scheduling, Time Tabling & Project Manage-ment (contributed)Contributed sessionChair: Ching Chih Tseng, Business Administration, DayehUniversity, 168 University Rd., Dacun, Taiwan, R.O.C., 51591,Changhua, Taiwan, cctseng@mail.dyu.edu.tw

1 - A heuristic approach for profit oriented disassemblysheduling problemMelike Kaya Akça, METU-Industrial Engineering Dept.,Graduate School of Natural and Applied Sciences, Turkey,kamelike@metu.edu.tr, Z. Pelin Bayindir, Ferda CanCetinkaya

9

MA-23 EURO-INFORMS - Rome 2013

We consider a disassembly scheduling problem for multiple productswith parts commonality and ample supply of discarded products. Wepropose a heuristic algorithm which is based on the idea of sequentiallysolving a relaxed version of the problem and single-period integer pro-gramming models. In a computational study, the performance of theheuristic approach is assessed. The results of the computational exper-iments indicate that the heuristic algorithm solves the problems in areasonable computational time and generates near optimal solutions.

2 - On an algorithm of multicriteria scheduling problemKetevan Kutkhashvili, Mathematics, University of Georgia,77 Kostava Street, 0102, Tbilisi, Georgia,kkutkhashvili@yahoo.com, Liana Qaralashvili

On the basis of general theory methods mathematical model is con-structed for such scheduling theory problems for which tasks imple-mentation is possible by single-step multiprocessing system in whichprocessors are half interchangeable and additional resources and setsof partial order are empty. The new multicriteria optimization algo-rithm (from the viewpoint of speed of acting and other certain criteria)is constructed.

3 - A bayesian approach to preventive maintenance op-timization in unreliable systemsDincer Goksuluk, Biostatistics, Hacettepe University, SıhhiyeCampus, Faculty of Medicine, Department of Biostatistics,Ankara - TURKEY, 06590, Ankara, Turkey,dincer.goksuluk@gmail.com, Selma Gürler, Deniz TürselEliiyi

In this study, we present a sequential preventive maintenance (PM)model for a component subject to failure. We assume an age reduc-tion model for imperfect PMs, performed whenever the component’sreliability falls below the threshold level, R. We consider the problemunder lack of data where Bayesian methods are more appropriate thanthe frequentist view. We model the total expected cost over all cy-cles, and the optimal preventive maintenance plan until replacementis found dependent on the threshold value. A numerical study is alsoconducted for sensitivity analysis of the developed model.

4 - Minimizing the total weighted completion time fortime-dependent jobsKrzysztof Ocetkiewicz, Department of Algorithms andSystem Modelling, Gdansk University of Technology, 11/12Gabriela Narutowicza Street, 80-233, Gdańsk, Poland,Krzysztof.Ocetkiewicz@eti.pg.gda.pl, Marek Kubale

We consider a single machine time-dependent scheduling problemwith linear deterioration and Sum wiCi objective. We present abranch-and-bound algorithm improved by elimination of dominatedpartial schedules. Such an algorithm makes possible to deal with hugememory requirements of the elimination while retaining running timespeedup. Also,it is capable of finding all Pareto optimal solutions sub-ject to Sum wiCi and Cmax criteria and scheduling jobs with prece-dence constraints. Our algorithm allows to double the size of instancesthat can be solved to optimality in a reasonable time.

� MA-23Monday, 8:30-10:00G6-3

Realistic Production Scheduling

Stream: Realistic Production SchedulingInvited sessionChair: Cyrus Siganporia, Biochemical Engineering, UniversityCollege London, United Kingdom, cyrus.siganporia.09@ucl.ac.uk

1 - New lower bounds for minimising the total comple-tion time in a two-machine non permutation flowshopscheduling problemSabrine Chalghoumi, Economie, Ecole supérieure dessciences économiques et commerciales de Tunis, 4, Rue AbouZakaria El Hafsi, 1089 Montfleury, Tunis, Tunisia,chalghoumisabrine@yahoo.fr, Mrad Mahdi, Talel Ladhari

The presented work addresses the two-machine non-permutation flowshop scheduling problem with the objective of minimizing the totalcompletion time subject to release dates. We proposed two new mixedinteger linear programming formulations. The first proposed MILPformulation is based on the Completion Time Variable formulationwhile the second one is based on a time indexed formulation. More-over, two new derived lower bounds are obtained by linear relaxationtechnique. Our computational experiments provide evidence that thesenew lower bounds present interesting and significant results.

2 - Solution Approaches For Flexible Job Shop Schedul-ing ProblemsSerife Aytug Balci, Industrial Engineering, Graduate Schoolof Natural and Applied Sciences, Ankara, Turkey,acelik@roketsan.com.tr, Meral Azizoglu, Cemal Oguzsoy

In this study, we consider FJSP existing in discrete parts manufactur-ing industries. We are motivated by the production environment of acompany, operating at Turkish defense industry. Our objective is tominimize the total weighted completion times of the jobs in the sys-tem. We formulate the problem as a mixed integer linear program andfind that our model could find optimal solutions only to small sizedproblem instances. For medium and large sized problem instances, wedevelop heuristic algorithms with high quality approximate solutionsin reasonable solution time.

3 - A genetic algorithm for flexible job shop schedulingwith overlapping in operationsYunus Demir, Industrial Engineering, Ataturk University,Erzurum Turkey, 25240, Erzurum, Outside US, Turkey,yunusdemir83@gmail.com, Selçuk Kürşat İşleyen

Flexible job shop scheduling problem (FJSP) is extension of job shopscheduling problem (JSP) which allows an operation to be performedby any machine among a set of available machines. In this paper wehandled FJSP with overlapping in operations. To solve this problema genetic algorithm (GA) is proposed. The proposed algorithm wastested on benchmark problems taken from literature of different scales.Obtained results were compared with the results obtained by other al-gorithms.

4 - Multi-objective optimisation of biopharmaceuticalproduction plans using a state task network repre-sentationCyrus Siganporia, Biochemical Engineering, UniversityCollege London, United Kingdom,cyrus.siganporia.09@ucl.ac.uk, Lazaros Papageorgiou,Suzanne Farid

Biotech firms face mounting pressure to meet demands for multiplecommercial therapeutics whilst minimising costs. Obtaining optimalcapacity plans across multiple sites is complicated by conflicting ob-jectives and products with batch and semi-continuous modes of pro-duction. This presentation describes development of a multi-objectiveSTN MILP model to help decide whether to outsource manufacturing,build a new facility, or do both as capacity limits are reached. An in-dustrial case study is presented showing how better long-term capacityplans can be generated when including various criteria.

� MA-24Monday, 8:30-10:00G6-4

Airport Operations Management

Stream: Project Management and SchedulingInvited sessionChair: Jeroen Belien, Center for Informatics, Modeling andSimulation, Hogeschool-Universiteit Brussel, Warmoesberg 26,1000, Brussels, Belgium, Jeroen.Belien@kuleuven.be

10

EURO-INFORMS - Rome 2013 MA-26

1 - A rolling horizon framework for an aircraft line main-tenance problemJorne Van den Bergh, HUBrussel, Belgium,jorne.vandenbergh@hubrussel.be, Jeroen Belien, Liesje DeBoeck, Erik DemeulemeesterWe present a rolling horizon framework for an aircraft line mainte-nance scheduling problem. The procedure starts with the creation ofa "robust’ base schedule in order to deal with the stochasticity in therelease times of the jobs. In the rolling horizon procedure, flight de-lays are simulated and we check whether the base schedule can handlethese delays with the available capacity and/or time buffers or whetherthe base schedule needs to be adjusted. When rescheduling is needed,we search for the cheapest solution that results in the smallest numberof changes to the base schedule.

2 - Flight and Crew Rescheduling in Post PerturbationSituationPeiman Alipour Sarvari, Industril engineering, GaziUniversity, Yunus Emre ogrenci yurdu, Altindag-Ulus, 06500,Ankara, Select, Turkey, peyman.alipour@gmail.comAbstract- Airline operations are disrupted frequently. The most impor-tant factors are unstable weather conditions and unpredictable techni-cal aeronautics problems. In a disrupted situation, an airline operatorneeds to quickly adjust the timetable and the resource schedules sothe importance of re-assignment of facilities to flight is much moreclearer. In this study, we considered the re-assignment of aircraft toflight after the emergence of disorder in the initial planning with regardto crew scheduling constraints. We propose a problem formulation anda heuristic solution algorithm.

3 - Aircraft Maintenance Scheduling and Optimizationfor Summer Period: A Case for Turkish AirlinesYavuz Ozdemir, Industrial Engineering, Yildiz TechnicalUniversity, Yildiz Teknik Universitesi Endustri MuhendisligiBolumu Barbaros Bulvari, Yildiz/Besiktas/Istanbul TURKEY,Istanbul, Turkey, Turkey, yavuzytu@gmail.com, Ahmet Bolat,Merve KaracaDecreased ticket prices in airline transportation make a competitiveairline market, and forces airline companies to make professional de-cisions for their activities. In this study, C type maintenance, whichis one of the important and long lasting maintenance, is assigned forsummer period considering the capacity of hangar with Microsoft Vi-sual Basic. Maximizing the utilization and minimizing the number orrequired aircrafts are considered as our aims. According to dynamicframe of airline operations and daily irregularities, manual editing isallowed in the code that studied in this paper.

� MA-25Monday, 8:30-10:00G9-1

Retail modelsStream: INFORMS MSOM StreamInvited sessionChair: Esra Cinar, Operations Management, IESE BusinessSchool, Cr. Arnús i Garí 3-7, 08034, Barcelona, Spain,Ecinar@iese.edu

1 - A closed-loop approach to dynamic assortment plan-ningEsra Cinar, Operations Management, IESE Business School,Cr. Arnús i Garí 3-7, 08034, Barcelona, Spain,Ecinar@iese.edu, Victor Martínez de AlbénizFirms offering short-lived products are constantly trying to keep thecustomers interested by refreshing their assortments, for example infashion retailing. We model these dynamics by assuming that productslose their attractiveness over time and we let the firm enhance the as-sortment at a cost, e.g., with new product introductions. We character-ize the optimal closed-loop policy that maximizes the revenue obtainedby increasing the attractiveness of an assortment, minus the associatedcost of the adjustments. Interestingly, we find that an assort-up-to pol-icy is optimal in some cases.

2 - Optimal Assortment Planning for Retailers UsingTransshipmentsNagihan Comez, Faculty of Business Administration, BilkentUniversity, Bilkent University Isletme Fakultesi MAZ 20,Bilkent, 06800, Ankara, -, Turkey, comez@bilkent.edu.tr,Alper Sen, Hilal Dag

Transshipments among retailers may help to recover stock-outs timelyand economically. In this study, our objective is to find the optimalassortments in a centralized retailer system, where retailers can utilizetransshipments for the products they don’t keep. Using an exogenousdemand model, we show that when all possible products can be rankedaccording to certain popularity criteria, the products in both the to-tal and common assortments of retailers will be chosen following thispopularity list at the optimality. We also show the benefits of trans-shipments wrt a no-transshipment system.

3 - Assortment Competition with Information RevelationVictor Martínez de Albéniz, IESE Business School,Barcelona, Spain, valbeniz@iese.edu, Gurhan Kok

In industries where customer needs quickly change, retailers can post-pone their assortment decisions (amount of variety, balance acrosscategories) if they are quick enough. We study here how assortmentcompetition depends on the postponement capabilities of retailers. Wedevelop a stylized model where two retailers choose their assortmentbreadth either before or after market characteristics are revealed.

� MA-26Monday, 8:30-10:00G9-7

Combinatorial Optimization: Applications

Stream: Combinatorial Optimization IInvited sessionChair: Jan van Vuuren, Department of Logistics, StellenboschUniversity, Private Bag X1, Matieland, 7602, Stellenbosch,Western Cape, South Africa, vuuren@sun.ac.za

1 - Choosing fuel treatment sites to reduce risk of fireJohn Hearne, Mathematical and Geospatial Sciences, RMITUniversity, GPO Box 2476v, 3001, Melbourne, Victoria,Australia, john.hearne@rmit.edu.au, James Minas

Reducing fuel loads is a method for reducing the risk of wildfires.Planning for this requires decisions relating to location, extent, andtiming of treatment. We formulate MIP problems to deal with this atthe landscape-level in the presence of resource limitations. We gener-ate a number of fuel treatment patterns and landscape metrics with aview to modifying the behaviour and effects of large wildfires. Meth-ods for incorporation of a range of ecological, operational and costconstraints are also considered.

2 - An exact dynamic programming algorithm for the air-craft sequencing problemMartin Kidd, Dipartimento di Elettronica, Informatica eSistemistica (DEIS), University of Bologna, Bologna, Italy,martin.kidd@unibo.it, Fabio Furini, Carlo Alfredo Persiani,Paolo Toth

We consider the static case of the aircraft sequencing problem on a sin-gle runway with constraints on the deviation from the first-come-first-serve order. An exact dynamic programming algorithm is proposed,and the effects of limiting the states in the state space graph are inves-tigated. Completion bounds are computed and are used together withheuristic upper bounds to further limit the number of states. For realworld air traffic instances from Milan Linate Airport it is shown thatthe algorithm is able to outperform the previously best exact algorithmfrom the literature.

11

MA-27 EURO-INFORMS - Rome 2013

3 - Finding bounds for the minimum number of movesfor solving the WrapSlide puzzleAlewyn Burger, Department of Logistics, StellenboschUniversity, Private Bag X1, Matieland, 7600, Stellenbosch,Western Cape, South Africa, apburger@sun.ac.zaWrapSlide is a slide-puzzle consisting of a 6 by 6 grid of tiles in whicheach quadrant of 3 by 3 tiles are coloured differently. The puzzle canbe scrambled by performing a number of moves involving wrappingof tiles. A move consists of sliding either the top, bottom, left or righttwo quadrants of tiles 1 to 5 units horizontally or vertically. We discussbounds on the minimum number of moves to unscramble the puzzle.

4 - A linear algorithm for secure domination in treesJan van Vuuren, Department of Logistics, StellenboschUniversity, Private Bag X1, Matieland, 7602, Stellenbosch,Western Cape, South Africa, vuuren@sun.ac.za, AlewynBurger, Anton de VilliersA secure dominating set of a graph is an assignment of guards to itsvertices, at most one guard per vertex, so that (a) every vertex is eitheroccupied by a guard or adjacent to an accupied vertex and (b) there is aguard in the neighbourhood of each unoccupied vertex who can moveto that vertex, again leaving a configuration satisfying (a) above. Inthis talk an algorithm is proposed for finding a secure dominating setof minimum cardinality for any tree in linear time and space.

� MA-27Monday, 8:30-10:00G9-8

Applications of Boolean Functions

Stream: Boolean and Pseudo-Boolean OptimizationInvited sessionChair: Endre Boros, RUTCOR, Rutgers University, 08854,Piscataway, New Jersey, United States,Endre.Boros@rutcor.rutgers.edu

1 - Discovery of strong association rules in binary ta-bles based on hypergraph dualizationKira Adaricheva, Mathematical Sciences, Yeshiva University,245 Lexington ave, 10016, New York, NY, United States,adariche@yu.eduDiscovery of (strong) association rules, or implications, is an importanttask in data management, that finds application in artificial intelligence,data mining and semantic web. We introduce the novel approach forthe discovery of a specific set of implications, called the D-basis, thatprovides a representation for a binary table, based on the structure ofits Galois lattice. At the core of the method is the D-relation definedin the lattice theory framework, and the hypergraph dualization algo-rithm that allows to effectively produce the set of transversals for agiven hypergraph.

2 - The performance of box-type classifiers in the logicalanalysis of numerical dataMartin Anthony, London School of Economics, WC2A 2AE,London, United Kingdom, M.Anthony@lse.ac.uk, JoelRatsabyThe use of boxes for pattern classification is a fairly natural way inwhich to partition data into different classes or categories. It is oneof the standard LAD approaches when dealing with numerical (ratherthan binary) data. We consider the performance of classifiers whichare based on unions of boxes. The classification method studied maybe described as follows: find boxes such that all points in the regionenclosed by each box are assumed to belong to the same category, andthen classify remaining points by considering their distances to theseboxes.

3 - Quadratic reformulations of nonlinear binary opti-mization problemsYves Crama, HEC - Management School, University ofLiège, Rue Louvrex 14 (N1), 4000, Liege, Belgium,Y.Crama@ulg.ac.be, Martin Anthony, Endre Boros, AritananGruber

We consider the problem of minimizing a pseudo-Boolean functionf(x), i.e., a real-valued function of 0-1 variables. Several authors haverecently proposed to reduce this problem to the quadratic case by ex-pressing f(x) as min g(x,y) s.t. y = 0, 1, where g is a quadratic functionof x and of additional binary variables y. We establish lower and up-per bounds on the number of additional y-variables needed in such areformulation, both for the general case and for the special case ofsymmetric functions like positive or negative monomials, k-out-of-nmajority functions, or parity functions.

4 - Threshold separability of boolean functions and itsapplication to data discretizationGiovanni Felici, Istituto di Analisi dei Sistemi ed Informatica,Consiglio Nazionale delle Ricerche, Viale Manzoni 30,00185, Roma, Italy, giovanni.felici@iasi.cnr.it, Endre Boros

We consider the conditions for the existence of a Linear ThresholdFunction that separates two sets of Boolean vectors obtained by dis-cretization of real valued data. We state a combinatorial necessary andsufficient condition for the existence of such function when points be-long to the plane, and show its equivalency with a straight-forwardinterpretation of the Farkas’ Lemma. Such equivalency provides inter-esting insights for our problem in larger dimensions and supports thedesign of a discretization algorithm, for which we provide computa-tional experience and comparative tests.

� MA-28Monday, 8:30-10:00G9-2

Geometric Clustering 1

Stream: Geometric ClusteringInvited sessionChair: Belen Martin-Barragan, Dpto. Estadistica, UniversidadCarlos III de Madrid, c/Madrid 126, 28903, Getafe, Madrid, Spain,belen.martin@uc3m.es

1 - On the existence of feasible power diagrams for op-timal weighted balanced clusteringsPeter Gritzmann, Mathematics, TU München, Arcisstr. 21,D-80290, Munich, Germany, gritzman@ma.tum.de, AndreasBrieden

A highly desirable property of a weighted clustering of points in d-space under balancing constraints is the existence of specific cell com-plexes that partition the space, and whose defining polyhedra con-tain the clusters, respectively. We introduce gravity polytopes andmore general gravity bodies that encode all feasible clusterings andshow that and how their extreme points correspond to strongly feasiblepower diagrams. Further, we characterize strongly feasible centroidalpower diagrams in terms of the local optima of some ellipsoidal func-tion over the gravity polytope.

2 - Soft power diagramsSteffen Borgwardt, Fakultät für Mathematik, TechnischeUniversität München, Boltzmannstr. 3, 85748, Garching,Bayern, Germany, borgwardt@ma.tum.de

In many applications, the points of a set in d-space are partitioned asa balanced least-squares assignment. Then there is a separating powerdiagram, for which each cluster lies in its own cell. In practice how-ever, data noise may destroy this nice property. We deal with this prob-lem by showing how to compute an optimal ’soft power diagram’ forgeneral data sets: These are power diagrams for which we are able tocontrol the number of points outside of their respective cluster’s cell.This leads to efficient algorithms for outlier detection and ’threshold-setting’.

12

EURO-INFORMS - Rome 2013 MA-30

3 - Approximation algorithm for plane polyhedral sepa-rability problemKonstantin Kobylkin, Optimization, IMM UB RAS, SophyaKovalevskaya st., 16, 620990, Ekaterinburg, Sverdlovskaya,Russian Federation, kobylkinks@gmail.com

We give an efficient approximation algorithm for an NP-hard sim-plest majority-based polyhedral separation problem for two sets on theplane which embeds minimum line covering problem as a special case.Specifically, we restrict separating lines to have positive constant termswhich makes the problem approximately solvable within constant rel-ative accuracy in some cases. Results of experiments will be given inthe talk on several problems of interest: polygon and star-like separa-ble integer-valued sets, mixtures of Gaussians, etc..

4 - Nested Variable Neighborhood Search for Classifica-tion and Clustering.Belen Martin-Barragan, Dpto. Estadistica, Universidad CarlosIII de Madrid, c/Madrid 126, 28903, Getafe, Madrid, Spain,belen.martin@uc3m.es, Emilio Carrizosa, Dolores RomeroMorales

Model selection is a critical issue in Data Analysis. Such selectioninvolves the choice of a suitable distance, or a kernel. Enriching themodels allows us to capture important information in the data, at theexpense of tuning more parameters. Existing metaheuristics are ef-fective when the dimension of the parameter space is low, but tend tofail when it is high. We customize a continuous VNS that is able tocope with complex and flexible models that involve a larger numberof parameters. We do it by exploiting the fact that rich models can beusually seen as generalization of simpler ones.

� MA-29Monday, 8:30-10:00G9-3

Management and control of queues

Stream: Stochastic Modeling / Applied ProbabilityInvited sessionChair: Dieter Claeys, Department TELIN, Ghent University,B9000, Ghent, Belgium, dclaeys@telin.ugent.beChair: Joris Walraevens, Department of Telecommunications andInformation Processing, Ghent University, B9000, Ghent,Belgium, jw@telin.ugent.be

1 - Performance evaluation of an overload control mech-anism in clustered servlet applicationsMarie-Ange Remiche, University of Namur, 5000, Namur,Belgium, mre@info.fundp.ac.be, Laurent Schumacher

Convergent Java Servlet infrastructure are key component in today’scluster computing. They support services for many different types ofprotocols and applications. However, to remain reliable, an admissioncontrol algorithm is needed. One is defined based on the actual loadof the servlet processor. A simple and efficient Markov model is buildup to measure the performance of this admission control mechanism.Performance measures of interest are then obtained by means of matrixanalytic technics. We validate our model in case of SIP traffic.

2 - Analysis and applications of coupled queuesEline De Cuypere, TELIN, University Ghent,St-Pietersnieuwstraat 41, 9000, Ghent, Belgium,eline.decuypere@telin.ugent.be, Koen De Turck, DieterFiems

A coupled queueing system consists of multiple queues, served by asingle server such that in every queue a customer leaves upon servicecompletion, and such that service is blocked if any of the queues isempty. Coupled queueing systems are frequently encountered in vari-ous contexts including assembly, production and communication net-works. While conceptually simple, the analysis and performance ofcoupled queues is a challenging task, which draws upon different anal-ysis methodologies. We focus on several numerical techniques anddevelop a novel Taylor series expansion.

3 - Size-aware MDP approach to dispatching problemsSamuli Aalto, Networking Laboratory, Helsinki University ofTechnology, P.O.Box 3000, FIN-02015, TKK, Finland,samuli.aalto@tkk.fi, Esa HyytiaDispatching problems are related to parallel server systems where newjobs arrive dynamically and each job is assigned to one of the serversupon its arrival. Assuming Poisson arrivals, any static dispatching pol-icy can, in principle, be improved by the policy iteration techniquedeveloped in the MDP theory. We summarize many recent studieswhere the state of the queueing system is described based on the sizeinformation of jobs. In addition, we give illustrative examples wherethese iterated policies are compared to well-known dynamic dispatch-ing policies (such as JSQ and LWL).

4 - A queueing model for group-screening facilitiesDieter Claeys, Ghent University, Belgium,Dieter.Claeys@telin.ugent.be, Joris Walraevens, HerwigBruneelClassification of items can often be achieved more economically byscreening the items in groups rather than individually. We model agroup screening facility by a batch-service queueing model where theservice (screening) time of a batch (group) is dependent on the numberof customers (items) within it. We calculate the fraction of time thescreening facility is screening and the average delay of items. Theseperformance measures enable to evaluate every possible screening pol-icy by defining the dependency between the service time of a batch andthe number of items within it appropriately.

� MA-30Monday, 8:30-10:00G9-10

Combinatorial Optimization &Applications

Stream: Discrete and Global OptimizationInvited sessionChair: Mikhail Goubko, Institute of Control Sciences RAS,Russian Federation, mgoubko@mail.ruChair: Xianping Guo, Zhongshan University, China,mcsgxp@mail.sysu.edu.cn

1 - Optimization of Cable Cycle: A Tradeoff between theSafety and CostBarsam Payvar, School of Technology and Business Studies,University of Dalarna, Plogstigen 20 C, 78451, Borlänge,Dalarna, Sweden, h09barpa@du.se, Hasan FleyehIn networks, length of the path is a factor which plays a central rolein the evaluation of the best path. In ring networks finding the short-est path is very hard. The problem is not a variation of the TravelingSalesman Problems (TSP) because the existence of Hamiltonian Cyclecannot be guaranteed. In such cases, the cycle may include more thanone route which means increasing risk of losing more nodes duringa failure. Thus, beside the cable length risk is another crucial factor.This work aims to optimize the network in order to find the shortestpath with minimum risk.

2 - Charging Station Allocation for Selfish Electric Vehi-clesTakashi Tsubota, Department of Industrial and SystemsEngineering, Aoyama Gakuin University,Chuo-ku,Fuchinobe,5-10-1, 2298558, Sagamihara,Kanagawa, Japan, c5612128@aoyama.jp, Shao-Chin SungWe are concerned with an allocation problem of charging stations forelectric vehicles (EV). Since EV’s batteries have limited capacities,plan of charging and delays caused by charging itself and by queuingfor charging have to be taken into consideration for routing of EVs.Hence, routing of EVs depends strongly on charging station alloca-tion. We propose a game theoretical approach based on an extendedmodel of selfish routing game, in which each charging station is con-sidered as a player whose location is decided based on its incentive,and the allocations are provided by stable solutions.

13

MA-31 EURO-INFORMS - Rome 2013

3 - On Communication Network Topology Problem withNode CostsMikhail Goubko, Institute of Control Sciences RAS, RussianFederation, mgoubko@mail.ruThe communication network topology problem is studied with no edgecosts and node costs depending on the node degree and capacity. If costdepends on the degree only, the optimal network is shown to be the treeT of known cost. If capacity only matters, a two-layered network N isoptimal where secondary nodes form a complete graph. When both de-gree and capacity matter, LB and UB are constructed using costs of Tand N networks. A branch-and-bound procedure is designed. The lo-cal search heuristics is suggested and tested on a collection of randomnetworks.

4 - Network Optimization Models for Resouce Allocationin Developing Military CountermeasuresBoaz Golany, Industrial Engineering & Management,Technion - Israel Institute of Technology, Technion City,32000, Haifa, Israel, golany@ie.technion.ac.il, Moshe Kress,Michal Penn, Uriel G. RothblumWe consider an arms race where an attacker introduces new weaponsto gain some advantage and a defender develops countermeasures thatmitigate the weapons’ effects. We address the defender’s problem:given limited resources, which countermeasures to develop and howmuch to invest in them to minimize the damage caused by the weaponsover a certain horizon. We formulate several models, corresponding todifferent operational settings as constrained shortest path problems andvariants thereof. We then demonstrate the potential applicability androbustness of our approach in various scenarios.

� MA-31Monday, 8:30-10:00G9-11

Retail Assortment Planning

Stream: Demand and Supply Planning in ConsumerGoods and RetailingInvited sessionChair: Heinrich Kuhn, Operations Management, CatholicUniversity of Eichstaett-Ingolstadt, Auf der Schanz 49, 85049,Ingolstadt, Bavaria, Germany, heinrich.kuhn@ku-eichstaett.de

1 - Assortment planning and inventory management un-der stock-out based substitution for perishable prod-uctsRob Broekmeulen, OPAC, TU Eindhoven, P.O. Box 516, Pav.E10, 5600 MB, Eindhoven, -, Netherlands,r.a.c.m.broekmeulen@tue.nl, Karel van DonselaarWe study a combined assortment planning and inventory optimizationproblem for perishable products in a retail environment with stock-outbased demand substitution. We develop a fast heuristic which enablesthe joint determination of optimal assortments and optimal reorder lev-els. We evaluated the impact of demand substitution on expected profit,outdating, assortment size and product availability. The impact of de-mand substitution on profit and outdating is large for product subcat-egories with a large assortment size, low total demand for the assort-ment, and small product lifetimes.

2 - Retail assortment, shelf space and inventory plan-ningAlexander Hübner, Operations Management, CatholicUniversity Eichstaett-Ingolstadt, Auf der Schanz 49, 85049,Ingolstadt, Germany, alexander.huebner@ku-eichstaett.de,Heinrich Kuhn, Sabine RaettigManaging assortment, shelf space and inventory is a core decision inretail, as increasing product variety is in conflict with the limited shelfspace. A retailer needs to define assortment size, assign shelf space anddefine reorder volume which then meets stochastic consumer demand.Our goal is to describe the underlying planning problem and developa Newsboy-model. We describe a capacitated multi-product assort-ment, shelf space and inventory management problem under stochasticdemand and substitution effects. Our numerical examples show signif-icant impact of substitution effects.

3 - Modeling Category Captaincy with Asymmetric Man-ufacturersUdatta Palekar, Business Administration, University ofIllinois at Urbana-Champaign, 350 Wohlers Hall, 1206 S.Sixth Street, 61820, Champaign, Illinois, United States,palekar@illinois.edu, Erik Bushey

Retailers are increasingly using category captains to manage multi-brand categories. We consider the asymmetric case where one brandhas more brand-loyals than other. We develop a game-theoretic modelto study the optimal strategy for the retailer and the two manufacturers.We present results that show when it is optimal for the retailer to selectthe smaller or larger manufacturer as a category captain. We find thatfor some situations it is not optimal for the retailer to have a categorycaptain. We also discuss the gains/losses incurred by each player dueto introduction of this category.

� MA-32Monday, 8:30-10:00G8-1

Supply Network Dynamics and DisruptionManagement

Stream: Supply Chain OptimizationInvited sessionChair: Dmitry Ivanov, Supply Chain Management, Berlin Schoolof Economics and Law, Germany, divanov@hwr-berlin.de

1 - Modeling and Analysis of Supply Chain ViabilityScott Mason, Clemson University, 124 Freeman Hall, 29634,Clemson, SC, United States, mason@clemson.edu, MariahMagagnotti, Kasin Ransikarbum

Supply chain viability is a network’s ability to grow, expand, recover,and/or develop. Today, the degree to which a supply chain pos-sesses these important characteristics that we believe facilitate multi-dimensional performance is unmeasured. We present our research fo-cusing on the question of how and where should organizations spendavailable resources in order to improve the viability of their supplychain? We motivate our research using a variety of different supplychain types: functioning supply chains, damaged/interrupted networks,and disturbed/re-organized chains.

2 - Disturbances and plan deviations: Process controlloops and quantitative analysisDmitry Ivanov, Supply Chain Management, Berlin School ofEconomics and Law, Germany, divanov@hwr-berlin.de, BorisSokolov

Integrated supply chain (SC) planning has been a visible research topicover the past two decades. Plan execution is subject to uncertainty.Disturbances may occur both in the flow/job shops and in the masterplanning environment and cause deviations in both schedules and mas-ter plans, influencing operative and tactical performance. Based on themutli-disciplinary theory of quantitative analysis and practical cases,this study presents models and corresponding control processes for thecontrol and adaptation of master plans and schedules in centralizedmulti-stage SCs.

3 - A State-space Model for Chemical ProductionSchedulingChristos Maravelias, Univeristy of Wisconsin - Madison,United States, maravelias@wisc.edu, James Rawlings