heur_gins.h
Go to the documentation of this file.
27 * @brief LNS heuristic that tries to delimit the search region to a neighborhood in the constraint graph
31 * Graph Induced Neighborhood Search (GINS) is a Large Neighborhood Search Heuristic that attempts to improve
32 * an incumbent solution by fixing a suitable percentage of integer variables to the incumbent and
35 * Its search neighborhoods are based on distances in a bipartite graph \f$G\f$ with the variables and constraints as nodes and
37 * Given an integer \f$k\f$, the \f$k\f$-neighborhood of a variable \f$v\f$ in \f$G\f$ is the set of variables, whose nodes
38 * are connected to \f$v\f$ by a path not longer than \f$2 \cdot k\f$. Intuitively, a judiciously chosen neighborhood size
41 * An initial variable selection is made by randomly sampling different neighborhoods across the whole main problem.
42 * The neighborhood that offers the largest potential for improvement is selected to become the local search neighborhood,
45 * GINS also supports a rolling horizon approach, during which several local neighborhoods are considered
46 * with increasing distance to the variable selected for the initial sub-problem. The rolling horizon approach ends
47 * if no improvement could be found or a sufficient part of the problem component variables has been part of
51 /*---+----1----+----2----+----3----+----4----+----5----+----6----+----7----+----8----+----9----+----0----+----1----+----2*/
Definition: struct_scip.h:68
type definitions for return codes for SCIP methods
type definitions for SCIP's main datastructure
common defines and data types used in all packages of SCIP
Definition: objbenders.h:43