Abstract:The no idle permutation flow shop scheduling problem (NIPFSP) is a popular NP-hard combinatorial optimization problem, which exists in several real world production processes. This study proposes a novel hybrid estimation of the distribution algorithm and cuckoo search (CS) algorithm (HEDA_CS) to solve the NIPFSP with the total tardiness criterion minimization. The problem model is built on the basis of the starting and ending time point of each job. A discrete solution representation method is applied in HEDA_CS to increase the operation efficiency. A novel probability matrix build method is also designed within the knowledge of the processing time matrix. The partially-mapped crossover operation works effectively during the CS phase. A suitable knowledge-based local search is also designed in the HEDA_CS to balance the exploitation and exploration. Finally, many simulations based on the new hard Ruiz benchmarks are conducted. Computational results demonstrate the effectiveness of the proposed HEDA_CS.