The Speakers

Ph.D. in Mathematics (UCM'97). Since 2016, Professor in Statistics and Operational Research. His research is focused on Mathematical Optimization as a basic discipline for modeling and solving problems in logistics and industrial engineering. In his more than 25 years of research experience, the following contributions can be highlighted. 1) BFC: Branch and Fix Coordination: general scheme for the resolution of Stochastic Optimization problems. 2) Air traffic flow management and conflict detection and resolution. 3) Risk management, with contributions to the measurement and management of risk in industrial planning problems in multi-stage problems.

He has participated in several research projects and in research contracts with companies and training activities in companies. He has more than 45 articles published in JCR journals, most of them in Q1 journals and several chapters in books edited by Springer, Wiley, and Mathematical Programming Society. Some of the articles have several hundred citations. He has more than two hundred presentations at international conferences and has given lectures at the Universities of Edinburgh, Chile, Humboldt, Seville, and the Polytechnic University of Barcelona, among others. He has actively participated in the organization of high-level scientific events and conferences. President of ALIO (Ibero-Latinoamerican Association of Operations Research) from January 2022.

Summary: This tutorial will introduce stochastic optimisation as a tool for optimal planning. Based on several examples, different approaches are introduced (as chance constraints, robust optimization, etc), to focus on Stochastic Optimisation via scenario analysis. As opposed to the traditional deterministic models, (in which all parameters are assumed to be known in advance), this approach represents the uncertainty in the problem by means of a representative (two- or multi-stage) scenario tree. We will analyse what kind of solutions are obtained and how to take advantage of the structure of the formulation to apply efficient decomposition methods (Langrange, Benders, etc.). In the last part of the presentation, we will deal with risk management and how to introduce risk aversion measures that allow us to obtain acceptable solutions for worst-case scenarios.

I received my degree at "Politecnico di Milano" where I did also my Ph.D. Then I did the post-doc at the Joint Research Centre (JRC) of the European Commission located in Ispra (Italie). I spent one year more in Ispra as a temporal agent of the European Commission. In 2001 I joined the LOSI team at the University of Troyes as Maître de conférences and at the same time, I was lecturer at Politecnico di Milano, INDACO Departement. In September 2008 I joined LIPN as a full professor of Operations Research. I animate the AOC team and I created and I animate GREFELOT (Groupe de REcherche Francilien sur l'Environement, la Logistique et le Transport). My main research interests are Combinatorial Optimisation, Mixed Integer Programming, Reformulation and Decomposition Methods, Vehicle Routing and Scheduling Problems, Design and Analysis of Combinatorial Algorithms, Metaheuristics, Reoptimization, Environmental Decision Support Systems. I'm interested in both Academic research as well Industrial applications.

Summary: Social networks are an integral part of social analysis because they play an important role in the spread of, e.g., information, innovation, or purchase decisions. A social network is defined as a graph with actors (or groups of actors) corresponding to nodes, and arcs corresponding to interactions between actors. Interactions may represent different concepts such as friendship, mentor-apprentice, one- or two-way communication, and so on. Recent years have witnessed a growing interest in the definition and study of mathematical models to represent the propagation of influence – broadly defined – in a social network, as well as in the identification of the actors that can play an important role in facilitating such propagation. Surprisingly this model fits perfectly also to model the spread of the COVID virus over a population.

The influence maximization problem is defined on a graph with an associated diffusion process that models the spread of influence on the graph. A node is defined as activated if it is affected by the diffusion process. A subset of the nodes are selected as seeds, and their role is to initialize the diffusion process. Influence propagates in a breadth-first manner starting from the seeds. Several rules can be used to model the activation of a node. An exact algorithm for the deterministic and robust IMP assuming a linear threshold model is presented in this talk. The proposed algorithm is based on a mixed-integer linear program (MILP) and Branch-and-Cut.

Víctor Blanco is an Associate Professor in Quantitative Methods for Economics & Business at the Universidad de Granada, Spain. He has a MSc in Pure Mathematics from the Universidad de Granada (2005) and PhD in OR from the Universidad de Sevilla (2009). His research has focused in a wide variety of topics related to Mathematical Optimization, such as Logistics & Transportation, Data Science, Computational Algebra or Algorithmic with more than 40 papers published in top OR journals, including Mathematical Programming, European Journal of Operations Research, Computers & Operations Research, Omega, Fuzzy Sets & Systems and Advances in Data Analysis and Classification. He has been interested both in the theoretical and applied aspects of Mathematical Programming. He is PI in several (local, regional, and national) research projects and has collaborated on different industrial contracts. He has carried several research stays at internationally recognized research centers like the Institute of Mathematics & its application (IMA, University of Minnesota), the University of British Columbia, the University of California-Davis and the University of Edinburgh. He is currently an active member of the Spanish Stats & OR Society (SEIO), the Spanish Network on Location Analysis and Related Problems (REDLOCA), and the Institute of Mathematics of the Universidad de Granada (IMAG).

Summary: Location Science (LS) is a very active research field in OR. The main goal of LS is to find the “optimal” position of one of more facilities in order to satisfy the requirements of a set of demand points. The nature of the facilites and the demand points, the way of measuring the goodness of the positions, the requirements for the facilities, the allocation patterns, the type of application, etc generate a wide amount of different challenging problems in this area. In this lecture I will provide an overview of different types of location problems and the different methodologies that have been developed to solve them. I will emphasize on the importance of Mathematical Optimization for solving Location problems, in particular integer linear programming and mixed integer second order cone programming. On the other hand, I will show the role of facility location in the design of novel tools in Machine Learning, concretely in the construction of classification and clustering rules, and how these optimization-based tools improve the existing methodologies.

Cipriano Santos professional goal is to investigate, invent, and develop novel applications of mathematical optimization techniques to increase the operational efficiency of business processes. During his 23 years at HP Labs, he participated on several applied research projects developing mathematical optimization models and decision support tools for Supply Chain Inventory Management, Customer Relationship Management, Optimal Resource Allocation for Large Data Centers, Large Scale Workforce Planning tools for the services industry –including tactical Resource Planning, Airline Dispatcher Workload Distribution Optimization, and Operating Room Planning and Scheduling. In January of 2018, he joined Gurobi Optimization as a Sr. Technical Content Manager. In September of 2021, Santos joined Tecnológico de Monterrey University as a Distinguished Professor of the school of Industrial Engineering. Santos has a bachelor degree in applied mathematics –Actuary with major in Operations Research, from the University of Mexico (UNAM), and a Master and PhD degrees in Operations Research from the University of Waterloo in Canada.

Summary: During this workshop we will discuss mathematical optimization (a.k.a. mathematical programming) using the Gurobi Python API. The workshop has two sessions of 2 hrs. each one. During the first session, we will provide an overview of the world of optimization and discuss mathematical optimization in the context of ‘Operations’ Artificial Intelligence. Then, we present the main components of a mathematical optimization model and discuss a simple manufacturing problem formulated as a linear programming model using the Gurobi Python API. Variations of the manufacturing problem will help us to explain various outcomes of a linear programming problem using the Gurobi Python API, such as a problem with multiple solutions, a problem that is unbounded, and a problem that is infeasible.

For the second session, we will present various ‘real’ life problems formulated as mixed integer linear programming models using the Gurobi Python API. Knowledge of Python at the beginners’ level is highly recommended. We will use Google Colab notebooks to formulate and run the mathematical optimization problems using the Gurobi Python API.

Juan G. Villegas holds a joint Ph.D. in Industrial Engineering/Systems Optimization between the Universidad de los Andes (Colombia) and the Université de Technologie de Troyes (France). He is a full professor of the Department of Industrial Engineering at Universidad de Antioquia (Medellín, Colombia).  He was the founding president of the Colombian Operational Research Society. Prof. Villegas also served as Secretary of ALIO (Asociación Latino Iberoamericana de Investigación Operativa) from 2019 to 2021. His research interests include the use of exact and metaheuristic methods for multi-objective optimization, facility location, and supply chain planning problems. Particularly, he is currently interested in the application of optimization models in the planning and performance evaluation of educational systems, agro-industrial applications, waste management, and the implementation of the circular economy.

Summary: The circular economy is an emerging concept that is gaining relevance in public policy and the academic literature. Defined as “an economic system that replaces the ‘end-of-life ’ concept with reducing, alternatively reusing, recycling and recovering materials in production/distribution and consumption processes. It operates at the micro-level (products, companies, consumers), meso-level (eco-industrial parks), and macro-level (city, region, nation, and beyond), with the aim to accomplish sustainable development, thus simultaneously creating environmental quality, economic prosperity, and social equity, to the benefit of current and future generations. It is enabled by novel business models and responsible consumers” (Kirchherr et al. 2017). The implementation of the circular economy concept within the operations of firms and supply chains encompasses several challenges, including the redesign of the products, the management of reverse/circular flows, and the definition of new business models (Agrawal et al, 2018).

Initially, this talk introduces the circular economy concept and some of the most salient strategies within it (e.g. industrial symbiosis, product-as-a-service, extended producer responsibility, etc.). Then, we review how different operations research tools can be applied to face these challenges by supporting the decision-making process in organizations, supply chains, and industries adopting the circular economy principle. Finally, we illustrate the application of operations research techniques to support decisions on the implementation of circular economy concepts on case studies arising in the Colombian context.

*joint work with Pablo A. Maya (Department of Industrial Engineering- Universidad de Antioquia-Colombia)