Summary: | 碩士 === 逢甲大學 === 統計學系統計與精算碩士班 === 107 === The two-stage three machines assembly scheduling problem has lots of applications in
industrial production management. The topic of learning effects (or truncation based)
has received growing attention in the field of scheduling. However, it is relatively
unexplored in the two-stage three machines assembly problem. In this study, we thus
investigated an assembly scheduling problem, with two machines in the first stage and
an assembly machine in the second stage, together with a truncated learning function.
The goal is to accomplish all jobs as soon as possible; i.e. to minimize the makespan.
Owing to the NP-hardiness of the proposed problem, we derive some dominance
propositions together with a lower bound for the branch-and-bound method for
finding the optimal solution. In additional, six versions of hybrids greedy iterated
algorithm, three versions of the local searches algorithm with and three versions
without a probability scheme, are proposed. The local search methods include a
pairwise interchange, an extraction and backward-shifted reinsertion, and an
extraction and forward-shifted reinsertion. At last, the performance results, and
statistical analysis of observations from the proposed algorithms are reported.
|