An effective adaptive iterated greedy algorithm for a cascaded flowshop joint scheduling problem

This paper addresses a novel scheduling problem, namely the cascaded flowshop joint scheduling problem (CFJSP), which has critical applications in the modern electronic information equipment manufacturing industry. The CFJSP is composed of a distributed permutation flowshop scheduling problem and hybrid flowshop scheduling problem. This paper considers how to arrange a variety of jobs for the two flowshops together to minimize total flowtime. We present a mixed integer linear programming mathematical model and an effective adaptive iterated greedy (AIG) algorithm with a decomposition and collaboration mechanism, which optimizes each production phase sequentially and ultimately optimizes the whole process. Combining the problem-specific characteristic, an adaptive inverse bounded heuristic, an adaptive bounded range local search, and an odd/even random insertion reconstruction mechanism are proposed to explore more valuable space. Comprehensive computational experiments and statistical analyses are conducted to verify the effectiveness of the proposed AIG. The experimental results show that the proposed AIG significantly outperforms the state-of-the-art competing algorithms from the literature in relative deviation index values at the same CPU running time.