A C++ Implementation of of Tabu Search for k-cardinality tree problem based on generic programming and component reuse

Abstract

We present a C++ implementation of Tabu Search method for the Minimum k-Cardinality Tree problem. Tabu Search is a well known heuristic for sub-optimally solving optimization problems. This method applies also to Minimum k-Cardinality Tree --an important problem to both theory and industrial applications. Our implementation is based on generic programming and re-usability of several components for Tabu Search method. We have compared our program with other ad hoc implementations for the problem and have observed a good performance of our implementation. Moreover, our implementation represents time savings, flexibility and robustness mainly due to the component reuse.