|
|
|
Plenary speakersThe following academics and industrials have accepted to give a plenary session during the workshop:
Stackelberg games and bilevel bilinear optimization problem Stackelberg Games confront contenders with opposed objectives, each wanting to optimize their rewards. Decision-making parties involve a party with the capacity of committing to a given action or strategy, referredto as the leader, and a party responding to the leader's action, called the follower. The objective of the game is for the leader to commit to a reward-maximizing strategy anticipating that the follower will best respond.
Finding the optimal mixed strategy of the leader in a Stackelberg Game is NP-hard when the leader facesone out of several followers and polynomial when there exists a single follower.
Games in which the strategies of the leader consist in covering a subset of at most K targets and the strategies of the followers consist in attacking some target are called Stackelberg Security Games and involve an exponential number of pure strategies for the leader. These games present many applications such as prevention of terrorist attacks, fare evasion in public transportation and illegal extraction of forest resources, regulation control of trucks, etc.
A Stackelberg game can be modeled as a bilevel bilinear optimization problem which can be reformulated as a single level mixed integer nonlinear program (MINLP). We present different reformulations of this MINLP and compare their LP relaxations from both theoretical and computational points of view.
Solving routing and scheduling problems through set-based modeling LocalSolver is a heuristic solver designed to practically tackle large-scale optimization problems. Having modeled your optimization problem using common mathematical operators, LocalSolver provides you with high-quality solutions in short running times. Combining different optimization techniques, LocalSolver scales up to millions of variables, running on basic computers. One of the strengths of LocalSolver is its rich yet simple modeling framework. Indeed, most usual mathematical operators are available, including arithmetical expressions (sum, product, exponential, logarithm, trigonometric functions) or logical expressions (and, or, comparisons, conditional terms, array indexing). As a consequence, there is no need to linearize the considered problem: the user can model it directly and naturally. Initially, this modeling power was based on numerical decision variables only (binary, integer, or floating-point). A significant extension to this approach was recently brought with the introduction of high-level structured decision variables, inspired from constraint programming set-based variables. Many optimization problems involve sequencing or ordering concepts: scheduling, routing, network design. For these problems, a new type of variables yields even simpler and more compact models. The value of such a variable is not a number but a collection of numbers. More precisely, a list variable list(n) represents a sub-permutation of the set {0,1,2,….n-1}. We will show in this presentation how this new kind of variables allows building very simple and very effective models for a number of optimization problems, including routing and scheduling problems.
Nowadays, with the increasing volatility of electricity prices and the growing importance of Demand Response (DR) programs, there are more and more incentives for energy-intensive industrial sites to be flexible in terms of electricity consumption and generation. However, the complexity of these large industrial plants, having to deal with many interconnected processes, multiple energy flows, integrated CHP and renewable electricity generation, as well as many technical constraints is often a barrier to fully leverage the different energy flexibility levers of the plant.
In this talk, we propose an advanced analytics approach to help industrial sites capture the full value of their energy flexibility. Thanks to integrated mathematical models of the full energy ecosystem of the plant which will include the constraints, the flexibility of the different processes and their complex interdependencies; it is now possible to predict the total cost impact of different energy management decisions and also to optimize these decisions with respect to electricity price and demand response incentives.
Based on this optimization model, key decisions can be taken over different time horizons:
We will illustrate this approach on steel, cement and pulp and paper industries.
Location and routing problems in modern telecommunication networks CANCELED DUE TO STRIKES FROM THE SNCF (FRENCH TRAINS) |
| Online user: 1 |
|
