Portrait of Andrea Lodi

Andrea Lodi

Associate Academic Member

Adjunct Professor, Polytechnique Montréal, Mathematics and Industrial Engineering Department

Founder and Scientific Director, IVADO Labs

Biography

Andrea Lodi is an adjunct professor in the Department of Mathematics and Industrial Engineering at Polytechnique Montréal, and founder and scientific director of IVADO Labs.

Since 2014, Lodi has held the Canada Excellence Research Chair in Data Science for Real-Time Decision-Making at Polytechnique Montréal. This is Canada’s leading research chair in the field of operations research.

Internationally recognized for his work on mixed linear and nonlinear programming, Lodi is focused on developing new models and algorithms to quickly and efficiently process massive amounts of data from multiple sources. These algorithms and models are expected to lead to the creation of optimized real-time decision-making strategies. The goal of his work as Chair is to apply this expertise in a range of sectors, including energy, transport, health, production and supply chain logistics management.

Lodi holds a PhD in systems engineering (2000) and is a full professor of operations research in the Department of Electrical, Electronic and Information Engineering at the University of Bologna. He coordinates large-scale European operations research projects, and has worked as a consultant for the CPLEX R&D team at IBM since 2006. Lodi has published over seventy articles in major journals in mathematical programming and also served as an associate editor for many of these journals.

His many honours include a 2010 Google Faculty Research Award and a 2011 IBM Faculty Award, and he was a member of the prestigious Herman Goldstine program at the IBM Thomas J. Watson Research Center in 2005–2006.

Publications

Predicting Tactical Solutions to Operational Planning Problems under Imperfect Information

Eric P. Larsen

Sébastien Lachapelle

Simon Lacoste-Julien

This paper offers a methodological contribution at the intersection of machine learning and operations research. Namely, we propose a method… (see more)ology to quickly predict expected tactical descriptions of operational solutions (TDOSs). The problem we address occurs in the context of two-stage stochastic programming, where the second stage is demanding computationally. We aim to predict at a high speed the expected TDOS associated with the second-stage problem, conditionally on the first-stage variables. This may be used in support of the solution to the overall two-stage problem by avoiding the online generation of multiple second-stage scenarios and solutions. We formulate the tactical prediction problem as a stochastic optimal prediction program, whose solution we approximate with supervised machine learning. The training data set consists of a large number of deterministic operational problems generated by controlled probabilistic sampling. The labels are computed based on solutions to these problems (solved independently and offline), employing appropriate aggregation and subselection methods to address uncertainty. Results on our motivating application on load planning for rail transportation show that deep learning models produce accurate predictions in very short computing time (milliseconds or less). The predictive accuracy is close to the lower bounds calculated based on sample average approximation of the stochastic prediction programs.

2018-07-31

ArXiv (preprint)

A polynomial algorithm for a continuous bilevel knapsack problem

Margarida Carvalho

Patrice Marcotte

2018-03-01

Operations Research Letters (published)

Existence of Nash Equilibria on Integer Programming Games

Margarida Carvalho

João Pedro Pedroso

2018-02-15

Springer Proceedings in Mathematics & Statistics (published)

Nash equilibria for integer programming games

Margarida Carvalho

João Pedro Pedroso

In this paper, we develop algorithmic approaches for a recently defined class of games, the integer programming games. Two general methods t… (see more)o approximate an equilibrium are presented and enhanced in order to improve their practical efficiency. Their performance is analysed through computational experiments in a knapsack game and a competitive lot-sizing game. To the best of our knowledge, this is the first time that equilibria computation methods for general integer programming games are build and computationally tested.

2018-01-01

(published)

www.semanticscholar.org

Nash equilibria in the two-player kidney exchange game

Margarida Carvalho

João Pedro Pedroso

Ana Luiza D'ávila Viana

2016-04-21

Mathematical Programming (published)

Experimental Algorithms

Samuel Rosat

Issmail ElHallaoui

François Soumis

2013-11-01

Lecture Notes in Computer Science (published)