How to start a heuristic? Utilizing lower bounds for solving the quadratic assignment problem

The Quadratic Assignment Problem (QAP) is one of the classical combinatorial optimization problems and known for its diverse applications. QAP an NP-hard problem which attracts use heuristic or metaheuristic algorithms that can find quality solutions in acceptable computation time. On other hand, there quite a broad spectrum mathematical programming techniques were developed finding lower bounds QAP. This paper presents fusion two approaches whereby from computations are used as starting points metaheuristic, called HC12, implemented on GPU CUDA platform. We perform extensive computational experiments demonstrate these bounding construction has significant impact resulting solutions.