How to solve real-world optimization problems with mixed-integer linear programming technology.
by prof Fanie Terblanche and dr Richard Oberdieck
Mixed-integer linear programming is not only an academic concern. It is currently adding value to thousands of companies across the globe by providing solutions to complex decision problems. Mixed-integer linear programming solutions are employed to solve strategic deci-sion-making, as well as operational planning problems.
In this tutorial, we will provide you with the opportunity to formulate mixed-integer linear programming problems within a web-based modeling environment and solve these pro-blems with Gurobi, the fastest mixed-integer programming solver in the world. For this pur-pose, attendees will make use of an internet browser – no installation of software or licenses will be required.
The following activities will form part of the tutorial agenda.
Getting acquainted with the modeling environment.
Making use of binary and integer decision variables for logical decision-making.
Formulating a well-known combinatorial optimization problem.
Introducing concepts of strong MIP formulations to improve computing performance.
Accessing and executing the optimization model from within Excel.
The trial accounts for the modeling environment will still be available for use by the attendees upon completion of the tutorial.
Iterated Local Search: Applications and Extensions
by prof Helena Ramalhinho Lourenço
Metaheuristics are general high-level procedures that coordinate heuristics and rules to find high-quality solutions to difficult optimization problems in a short computational time. They are designed to solve large-scale complex optimization problems that cannot be solved in reasonable processing time by the classic combinatorial optimization methods. Metaheuristics have been extensively applied to solve real problems in many areas from transportation to finance, sports or manufacturing, etc..
In this tutorial, we will briefly review several metaheuristics successfully applied to combinatorial optimization problems, and we will focus on the Iterated Local Search (ILS) approach, a conceptually simple and efficient well-known metaheuristic. The main idea behind ILS is to drive the search not on the full space of all candidate solutions but on the solutions that are returned by some underlying algorithm; typically, local optimal solutions obtained by the application of a local search heuristic. This method has been applied to many different optimization problems. We will review briefly the metaheuristics ILS method and describe two relevant extensions: the hybrid ILS approach that combines ILS with other metaheuristics and/or exact methods (MathILS) and the SimILS, that combines Simulation with ILS, to solve Stochastic Combinatorial Optimization Problems. We will discuss the advantages and disadvantages of these extensions and present some real applications in areas like Supply Chain Management, Logistics, Production, Marketing and Health and Social Care.