OPTIMA seminars

Upcoming events

From September 2014, the seminars are jointly organized by Dolphin and OSL, both members of the OPTIMA thematic group from the CRIStAL lab.

  • Wednesday, November 30, 2016 (11:00, room B31, Inria)
    Daniele Vigo (Professor, University of Bologna) : Visual Attractiveness in Routing Problems
    Enhancing visual attractiveness or “beauty” in a routing plan has proven to be an effective way to facilitate practical implementation and positive collaboration among tactical and operational levels in the organization. This is the reason why many authors have considered, either explicitly or implicitly, such aspect in the optimization process for different routing applications. However, due to its subjective nature, there is not a single way of evaluating visual beauty of a routing solution. The aim of this talk is to provide an overview of the literature on visual attractiveness and to show the results of a heuristic algorithm incorporating explicitly visual attractiveness optimization. In particular, we analyse the different metrics that were used to model the visual attractiveness of a routing plan and provide guidelines that planners and researchers can use to select the method that better suits their needs.
    (joint work with Diego Gabriel Rossit, Fernando Tohmé, Mariano Frutos).

Past events

  • Thursday, November 3, 2016 (14:00, room B21, Inria)
    Véronique François (Teaching Assistant, HEC Liège) : Neighborhood search approaches for the multi-trip vehicle routing problem with time windows We introduce two large neighborhood search approaches for solving the multi-trip vehicle routing problem with time windows, where each vehicle can perform several routes to serve a set of customers. Besides considering a time window at the depot, the working time of each vehicle may not exceed a maximum duration, while its start time is a decision variable. We compare specific operators that tackle the routing and the assignment aspects of the problem simultaneously with a method combining bin packing techniques and routing heuristics. An automatic configuration tool is used, not only to improve the quality of the results, but also as a mean to gain knowledge about algorithm design options and their interactions. This enables us to question the impact of commonly used heuristic components. (joint work with Y. Arda and Y. Crama)
  • Tuesday, October 11, 2016 (15:00, room B21, Inria)
    Miguel Anjos (Professor at Polytechnique Montréal) : Pairwise influences in dynamic choice: theories, models and methods

Choices of different individuals over time exhibit pairwise associations in a wide range of economic contexts. Network models for the processes by which ideas and decisions propagate through social interaction have been studied in different streams of literature. However, when the network structure is unknown, only few contributions have focus on detecting the underlying pattern of pairwise interaction from sequences of multidimensional decisions. In fact, while it is typically possible to directly observe individual choices, inferring individual influences (who influences who) might be difficult in the general case, as it requires strong modeling assumptions on the cross-section dependencies of the associated multidimensional panels. We provide an overview of the existing approaches and investigate a class of exponential random models to jointly deal with dynamic choices of individuals over items together with the structure of pairwise influences between them. We argue that computational problems, related to the intractability of the normalizing constant, emerge when estimating the unknown parameters of this class of models. This drawback can be overcome by embedding the defined model into a Bayesian estimation framework and applying a specialized MCMC procedure, based on the joint simulation both from the parameter and the sample spaces. After a detailed analysis of the proposed statistical methodology, we present a practical application to a data set of songs, that are diffused over radio and TV stations; we infer station-to-station influences. This allows solving the classical influence propagation problem and take decision over the best stations in which one should first launch a song to guarantee maximum propagation.

  • Monday, October 10, 2016 (14:00, room B21, Inria)
    Miguel Anjos (Professor at Polytechnique Montréal) : Smart Grids and Optimization: A Winning Combination

A smart grid is the combination of a traditional electrical power system with information and energy both flowing back and forth between suppliers and consumers. This new paradigm introduces major challenges such as the integration of intermittent generation and storage, and the need for electricity consumers to play an active role in the operations of the system. We will summarize the opportunities provided by smart grid to the optimization community, and illustrate one such opportunity through some recent research on optimal aggregation of energy resources(joint work with F. Gilbert, P. Marcotte, and G. Savard).

  • Tuesday, March 1, 2016 (14:00, room B31, Inria)
    Diego Cattaruzza (Maître de Conférences, Centrale Lille): Vehicle Routing Problems for City Logistics

City Logistics has been introduced to study and understand the transport activities in urban areas. Optimisation of urban delivery planning needs to take into account the dynamics of the city in order to provide efficient services. In nowadays delivery systems it is common to forbid heavy trucks to enter city centers. Trucks are forced to unload at logistic platforms located in the outskirts of the city. Merchandise is then loaded into smaller (and possibly eco-friendly) vehicles in charge of final delivery to customers. First we present a survey of the movements of goods that occur in cities. This motivates the study we conducted and the development of optimisation tools for delivery activities in urban areas. Second, we introduce the Multi-Trip Vehicle Routing Problem with Time Windows and Release Dates. Routing problems that take into account time windows and multiple delivery trips are commonly studied by the scientific community, especially when applied to the urban context. On the other side, the study of routing problems that consider release dates on the products is not. Here the release date models the instant the merchandise is available at the logistic platform and ready for final dispatching. A genetic procedure is presented to tackle the problem.\\This work was conducted in the context of the MODUM project (http://www-lipn.univ-paris13.fr/modum) that was founded by the French National Research Agency - ANR.

* Monday, November 23, 2015 XXX, February XX, 2016 (13:30, room B11, Inria)\\Luis Paquete (University of Coimbra, Portugal, invited Professor in Dolphin): Solution set compression in multiobjective optimization\\A major drawback of implicit enumeration algorithms for multiobjective combinatorial optimization problems is the large usage of memory resources that is required to store the set of potential solutions during the search process. We introduce several techniques that allow to compress a set of solutions during the run of a dynamic programming algorithm for the particular case of the biobjective knapsack problem. Particular emphasis is given on understanding the trade-off between memory usage and computation time, both from a theoretical and practical point of view. The experimental results indicate that some of these techniques allow to have a high compression ratio with very small computational time overhead.

  • Thursday, February 4, 2016 (14:30, room B11, Inria)
    Pierre-Louis Poirion (Post-doc at LIX, Ecole Polytechnique): Optimization of dynamical systems, bilevel programming and observability on smart-grids

In the french "Sogrid" project about smart-grids, our aim is to place the minimum number of measurement devices on the links of a power-grid in order to "observe" to whole grid. After studying the complexity of the problem, we present a first natural but computationally unsatisfactory MILP model, which we reformulate into a bilevel program using a fixed point argument. We will then study a large class of dynamical systems depending on parameters which we want to optimize. We will show that under some dominance hypothesis, these problems can be reformulated into conic bilevel programs. We then present a new constraints and variables generation algorithm to solve the problem and show that the observability problem can be solved by a simplified version of the algorithm. Finally we discuss some relations between the observability problem and the minimum rank of a graph and show that some bounds can be found to speed up the algorithm.

  • Wednesday, December 09, 2015 (14:00, room B21, Inria)
    Myriam Delgado (Federal University of Technology, Paraná, Brazil, invited Professor in Dolphin): METAHEURISTICS: from Optimization to MetaLearning
  • Thursday, September 24, 2015 (13:30, room B21, Inria)
    Khadidja Yachba (PhD student, University of Oran, Algeria): Gestion du problème de placement de conteneurs dans le transport maritime de marchandises en utilisant une méthode multicritère
  • Thursday, September 24, 2015 (14:00, room B21, Inria)
    Alex Freitas (University of Kent, UK): Hierarchical feature selection for the classification of ageing-related genes
  • Wednesday, September 9, 2015 (14:00, room B11, Inria)
    - Aymeric Blot : Multi objective ParamILS
    - Benjamin Fisset : Bilan ADT MoMine
  • Tuesday, September 8, 2015 (15:00, amphi Turing, M3)
    Hernan Aguirre (Shinshu University, Nagano, Japan): Evolutionary Many-objective Optimization
  • Tuesday, June 23, 2015 (10:00, room B21, Inria)
    - Greet Vanden Berghe (CODeS, KU Leuven, Belgium): Nurse rostering, academic challenges
    - Patrick De Causmaecker (CODeS, KU Leuven, Belgium): Algorithm engineering
    - Tony Wauters (CODeS, KU Leuven, Belgium): The Traveling Umpire Problem
  • Tuesday, June 9, 2015 (13:30, room C310, Ecole Centrale de Lille)
    Lijie Bai (PhD student, Ecole Centrale de Lille): Trains platforming problem in busy and complex railway stations
  • Thursday, April 16, 2015 (10:30, room C310, Ecole Centrale de Lille)
    Anne Quesnel-Barbet (Centre Hospitalier Régional et Universitaire de Lille): Operations research in the fields of health and society
  • Wednesday, April 8, 2015 (13:30, room B21, Inria)
    Fabio D'Andreagiovanni (Dept. of Mathematical Optimization, Zuse Institute Berlin (ZIB), Berlin, Germany) : Multiband Robust Optimization: theory and applications
  • Tuesday, March 17 March, 2015 (10:00, salle plénière - A building, Inria)
    H. J. Siegel (Colorado State University, USA): Stochastically Robust Resource Management for Heterogeneous Computing Systems
  • Thursday, February 19, 2015 (14:00, room B31, Inria)
    Qingfu Zhang (City University of Hong Kong, Hong Kong / University of Essex, UK): Combination of Evolutionary Algorithms with Experimental Design, Traditional Optimization and Machine Learning
  • Tuesday, January 20, 2015 (11:00, room B21, Inria)
    Andrei Tchernykh (Computer Science Department, CICESE, Ensenada, BC, México): Understanding Uncertainty and Scheduling in Grid/Cloud environment
  • January 07, 2015 (14:00, room C310, Ecole Centrale de Lille)
    Nelson Maculan (Federal University of Rio de Janeiro, Brazil): Two Applications of Mixed Integer Nonlinear Programming (MINLP) in R^n: the Euclidean Steiner Tree Problem and Covering a Solid with Different Spheres
  • December 03, 2014 (10:30, room B11, Inria)
    Bernard Gendron (Univ. Montreal, Canada): Piecewise Linear Multicommodity Flow Problems
  • November 20, 2014 (16:00, room B11, Inria)
    Bayrem Tounsi: Programmation mathématique avec contrainte d'équilibre stochastique pour un problème de tarification optimale de services de livraison
  • October 22, 2014 (14:00, room B11, Inria)
    Juan Jose Palacios: Metaheuristic Strategies for fuzzy Scheduling problems
  • October 15, 2014 (14:00, room C310, Ecole Centrale de Lille)
    Maxime Ogier: Network design for short/local supply chain and decentralized planning (joint work with V-D. Cung and J. Boissière)
  • January 23, 2013 (14:00, room B11)
    Philippe Nghe (Lab. of Biochemistry, ESPCI ParisTech): Pareto front of gene regulatory networks
  • September 19, 2013 (14:00, room B11)
    Trong-Tuan Vu: Adaptive Dynamic Load Balancing in Heterogenous Multiple GPUs-CPUs Distributed Setting: Case Study of Branch-and-Bound Tree Search
  • 28/05/2013 (10:00, room B11)
    Nicolas Zufferey (HEC – University of Geneva): Learning Tabu Search for Combinatorial Optimization
  • 14/05/2013 (11:00, room B11)
    Ekaterina Alekseeva: A matheuristic for the discrete (r|p)-centroid problem with multiple objectives for the followerslides ]
  • 05/04/2013 (13:00, room B11)
    Sezin Afşar: A Bilevel Approach to Energy Pricing Problem with Smart Grids
  • 05/03/2013 (17h00, salle B11)
    Isam Shahrour: Projet SUNRISE - smart grids
  • 05/02/2013 (13h45, salle B11)
    Dimo Brockhoff: Presentation of the COCO platform
  • 22/01/2013 (15h00, bâtiment M3, amphi Turing)
    Patrick De Causmaecker (Faculty of Engineering of the Katholieke Universiteit Leuven, Campus Kortrijk): Pareto Optimal Negotiation for Nurse Rostering Problems [ slides ]
  • 13/12/2012 (13h45, salle B11)
    Trong-Tuan Vu: Dynamic Load Balancing: Design and Large Scale Experimentsslides ]
  • 04/12/2012 (14h, salle W01, Bat A)
    Daniele Catanzaro (Université Libre de Bruxelles): The loss of heterozygosity problem
  • 15/11/2012 (13h45, salle B11)
    Sophie Jacquin: Optimisation des profits dans la production d’énergie hydraulique
  • 25/10/2012 (11h, salle B11)
    Rita Alexandra Santos Gonçalves Macedo (Post-doctoral CNRS, Projet CISIT, LAMIH, Université de Valenciennes et du Hainaut-Cambrésis): Network-flow pseudo-polynomial models
  • 20/09/2012 (13h45, salle W01)
    Aline Parreau: "Problèmes d'identification dans les graphes" [ slides ]
  • 18/06/2012 (11h, salle W21)
    Gleb Belov (University Duisburg-Essen): "Consecutive-ones matrices and LP for packing and scheduling"
  • 06/04/2012 (13h45, salle W11)
    Julie Jacques:"Datamining and optimization methods to improve recruiting in clinical trials" [ slides ]
  • 13/03/2012 (13h45, salle W21)
    Marie-Emilie Voge
  • 16/02/2012 (13h30, salle W01)
    Julie Hamon:"Coopération entre optimisation combinatoire et statistiques pour la sélection animale" [ slides ]
  • 08/02/2012 (14h, salle du conseil)
    Nicolas Dupin:"Optimisation robuste"
  • 07/02/2012 (15h, salle du conseil)
    Nicolas Dupin: "Scheduling nuclear plants outages: Mixing exact and heuristic methods"
  • 26/01/2012 (13h45, salle W21)
    Florian Richoux (Université de Tokyo):"Experiments in Parallel Constraint-Based Local Search"
  • 05/01/2012 (13h30, salle W21)
    Sébastien Verel:"Paysage de fitness, optimisation multiobjectif, et contrôle de paramètres" [ slides ]
  • 04/11/2011 (14h, salle du conseil)
    - Gaël Even (Gènes Diffusion) : "Les problématiques de la sélection animale"
    - Julie Hamon (Dolphin/Modal, thèse CIFRE Gènes Diffusion) : "Sélection de variables en régression pour la génétique animale"
    - Laetitia Jourdan (Dolphin) : "Méthode d'optimisation combinatoire pour le bi-clustering"
    - Christophe Biernacki (Modal) : "Modèle probabiliste pour le bi-clustering"
  • 06/10/2011 (13h45, salle W21)
    Daniel Porumbel (LGI2A, Univ. Artois): "Une perspective algorithmique sur trois problèmes discrets: isomorphisme, coloration de graphe et distance entre partitions"
  • 24/06/2011 (13h45, salle du conseil)
    Mostepha Redouane Khouadjia: "Metaheuristics for Solving Dynamic Vehicle Routing" [ slides ]
  • 23/06/2011 (10h, salle W21)
    Frédéric Saubion (Univ. Angers): "Solveurs autonomes"
  • 10/06/2011 (13h45, salle du conseil)
    Thé Van Luong: "A Thorough Examination of Metaheuristics on GPU" [ slides ]
  • 01/04/2011 (13h45, salle du conseil)
    Fabio Daolio: "Local-optima-network modeling of hard combinatorial landscapes" [ slides ]
  • 25/03/2011 (13h45, salle du conseil)
    Marie-Éléonore Marmion: "Using neutrality to solve optimisation problem" [ slides ]

Abstract. Solving efficiently complex problems using metaheuristics, and in particular local search algorithms, requires incorporating knowledge about the problem to solve. It is well known that in such problems, several solutions may have the same fitness value. As this neutrality property is an important issue, it should be taken into account during the design of search methods. The permutation flow shop problem (FSP) has this property. Then, a deep landscape analysis focused on the neutrality property is driven and propositions on the way to use this neutrality in order to guide the search efficiently are given: 1) NILS (Neutrality-based Iterated Local Search) allows neutral walks to move on the plateaus. This method has been experimented on the FSP and we show that NILS is able to find improving solutions compared with a classical iterated local search. The tradeoff between the exploitation of neutrality and the exploration of new parts of the search space are also analyzed. 2) VEGAS (Varying Evolvability-Guided Adaptive Search) that escapes from plateaus considering the whole neutral network rather than an arbitrary solution. Preliminary experiments have been performed on NK-landscapes with neutrality. Results show the importance of considering the whole neutral network and of guiding the search cleverly. Solving FSP with VEGAS approach should give good performance.