Quantum Simulated Annealing Algorithm

Received on:2/6/2009 Accepted on:1/4/2010 Abstract Simulated annealing (SA) has been considered as a good tool for search and optimization problems which represent the abstraction of obtaining the crystalline structure through a physical process. This algorithm works sequentially that the current state will produce only one next state. That will make the search to be slower and the important drawback is that the search may fall in local minimum which represent the best solution in only part of the solution space. In this work we present the transformation of Simulated Annealing algorithm into quantum version which will be called Quantum Simulated Annealing (QSA). This algorithm will overcome the drawbacks of slowness and local minimum falling by produce as much as possible of the neighbor states and work on in parallel by exploiting the massive parallelism feature in quantum computation. The results show that QSA can find the optimal path in smaller number of iterations than the sequential simulated annealing algorithm and the time complexity of QSA is better than any other parallel simulated annealing algorithm.