Keynote speakers

Keynote speaker 1 :  Prof. Juan-Jose Salazar-Gonzalez (Universidad de La Laguna, Tenerife, Spain)

Title:  Travelling Salesman Problems with Time Consistency

Abstract: A typical requirement in Scheduling is the need of synchronizing several machines to jointly perform each task. Similarly, in Logistic, there may be the need of synchronizing several vehicles to jointly serve a customer. In most articles, this requirement is modelled with continuous variables and big-M values in formulations with poor linear programming relaxations.

In this talk, we will show an alternative Combinatorial Optimization approach based on computing paths (special network flows). We describe the approach on some variants of the well-known Travelling Salesman Problem. In both variants, there is a time period (say, a week) and customers asking for a visit on some specific days of the week.

The aim in both variants is to design a route for each day such that, when a customer requires service in two days, the service time is "similar" in both days. We will discuss several mathematical formulations for the variants and analyse computational results comparing computer implementations. While the approach based on computing paths leads to models with a large number of variables, appropriated Bender' Decomposition techniques make the approach be the winner in our experiments.

The talk will present other problem variants where the new approach (or others) could lead to interesting and useful results in Combinatorial Optimization. The content of the talk is partially based on the article https://doi.org/10.1016/j.ejor.2023.08.021, EJOR (2024) Volume 313, Issue 2, Pages 465-477.

Keynote speaker 2 :  Prof. Suresh Sethi (University of Texas, Dallas, USA)

Title:    Integer Programming Approaches to Forecast Horizons in Dynamic Lot-Size Problems and Extensions

Abstract:  For most multi-period decision-making problems, the influence of information about later periods on the optimal decision in the current period reduces as we move farther into the future. If and when this influence reduces to zero, we refer to the corresponding problem horizon as a forecast horizon. For real businesses, the problem of obtaining a minimal forecast horizon becomes relevant because estimating reliable data for future periods gets progressively more challenging and expensive. We present structural and computational investigations of a class of weak forecast horizons—minimal forecast horizons assuming that future demands are multiples of a given positive integer—for a specific class of dynamic lot-size (DLS) problems. Apart from being appropriate in most practical instances, the discreteness assumption offers a significant reduction in the length of a minimal forecast horizon over the one using the classical notion of continuous future demands. The discreteness assumption allows us to characterize forecast horizons as feasibility/optimality questions in 0-1 mixed-integer programs. On an extensive test bed, we demonstrate the computational tractability of the integer programming approach.

Keynote speaker 3 :  Prof. Luis Gouveia (Department of Statistics and Operations Research, Faculty of Sciences of the University of Lisbon, Centre of Mathematical Studies of the University of Lisbon CEMS-UL)

Title:  Models for the Hamiltonian p-Median Problem

Abstract:  Given a positive integer p and a weighted undirected graph G=(V,E), the Hamiltonian p-Median Problem (HpMP) on G is to find a minimum weight set of p elementary cycles partitioning V. Mixed Integer Linear Programming (MILP) formulations for this problem are often based on an assignment formulation, for which the set of feasible solutions corresponds to the set of any number of node-disjoint cycles covering the nodes of the graph, extended with two sets of constraints preventing solutions with more and less than p cycles, respectively.

In this session, we begin by discussing different constraints to prevent more than p cycles. Such constraints are often generalizations of subtour elimination constraints already known for the Traveling Salesman Problem (TSP) (note that the TSP is a HpMP with p = 1). Thus, there exists a vast knowledge othat can be used to derive inequalities to prevemt more than p cycles. Even then, there are still interesting and nonobvious results concerning these constraints.

In contrast, constraints to prevent less than p cycles are much less well known. This lack of knowledge is also evidenced by computational results in the literature on the HpMP for large values of p where these constrainbst are nore effective.

In the second part, we focus on sucvh sets of constraints. Both extended (compact) and natural formulations (including an exponentialy sized set of quasi hamiltonian cycle constraints) are discussed. Finally, future research directions are proposed.

Collaborations with Michele Barbato, Francisco Canas and Pierre Pesneau

Keynote speaker 4 :  Dr. Julien DARLAY (Hexaly, France)

Title:  Hexaly, a new kind of global optimization solver

Abstract: Hexaly Optimizer is a new kind of global optimization solver. Its modeling interface is nonlinear and set-oriented. In a sense, Hexaly APIs unify modeling concepts from mixed-linear programming, nonlinear programming, and constraint programming. Under the hood, Hexaly combines various exact and heuristic optimization methods: branch-and-bound, automatic Dantzig-Wolfe reformulation, column and row generation, propagation methods, local search, direct search, and surrogate modeling techniques.

Regarding performance benchmarks, Hexaly distinguishes itself against the leading solvers in the market, like Gurobi, IBM Cplex, and Google OR Tools, by delivering fast and scalable solutions to Routing, Scheduling, Packing, Clustering, and Location problems.

This talk will introduce our set-based modeling formalism and show its scalability for large instances. We will then explore how the solver can use this formalism to automatically use state-of-the-art resolution techniques from the exact and heuristic fields.

Author: Julien DARLAY

Julien obtained a Ph.D. in computer science from Grenoble University (2011). His research was at the intersection of mathematical optimization and machine learning. He was awarded in several international OR competitions: 2nd Junior Prize of ROADEF 2009 Challenge, 3rd Senior Prize of ROADEF 2010 Challenge, Kaggle Santa’s Stolen Sleigh 2015 Challenge. Julien is now Head of Science at Hexaly.