In the M.Sc. thesis, I investigated the blocking flow scheduling problem for minimizing the total completion time. The blocking flow shop is a flow shop without any buffer between stages. I developed different exact and heuristic methods for solving this problem.
First, I proposed an iterated greedy algorithm. This algorithm found new best solutions for 88 out of 120 benchmark instances.
Second, I developed a branch and bound algorithm. Some lower bound procedures based on the graph for a specific sequence, and also some dominance rules were applied in this B&B algorithm. This algorithm found the optimal solutions of 17 instances of Taillard instances in less than 20 minutes.
After defending my thesis and before starting my PhD, I also investigated the limited-buffer flow shop scheduling problem with makespan criterion. The blocking flow shop is a special case of the limited-buffer flow shop. For this problem, I developed a speed-up method based on the graph of a sequence for increasing the efficiency of local search approaches. I presented a hybrid VNS algorithm that was superior than state-of-the-art algorithms.