Feasibility Pump-like heuristics for mixed integer problems

Authors:
M. De Santis;S. Lucidi;F. Rinaldi
Affiliations:
-;-;-
Venue:
Discrete Applied Mathematics
Year:
2014

Citing 11
Cited 0

A heuristic ceiling point algorithm for general integer linear programming

Management Science
Tabu Search

Tabu Search
General Purpose Heuristics for Integer Programming—Part II

Journal of Heuristics
Exploring relaxation induced neighborhoods to improve MIP solutions

Mathematical Programming: Series A and B
The feasibility pump

Mathematical Programming: Series A and B
Pivot, Cut, and Dive: a heuristic for 0-1 mixed integer programming

Journal of Heuristics
Concave programming for minimizing the zero-norm over polyhedral sets

Computational Optimization and Applications
MIPLIB 2003

Operations Research Letters
A feasibility pump heuristic for general mixed-integer problems

Discrete Optimization
Improving the feasibility pump

Discrete Optimization
Pivot and shift-a mixed integer programming heuristic

Discrete Optimization

Quantified Score

Hi-index	0.04

Visualization

Abstract

Finding a feasible solution to a MIP problem is a tough task that has received much attention in the last decades. The Feasibility Pump (FP) is a heuristic for finding feasible solutions to MIP problems that has encountered a lot of success as it is very efficient also when dealing with very difficult instances. In this work, we show that the FP heuristic for general MIP problems can be seen as the Frank-Wolfe method applied to a concave nonsmooth problem. Starting from this equivalence, we propose concave non-differentiable penalty functions for measuring solution integrality that can be integrated in the FP approach.