<p>The manufacturing of complex products requires more than one workshop. As a result, regarding the scheduling problem, how to assign the job processing sequence in heterogeneous workshops is a significant consideration. Meanwhile, it is completely different from the single workshop scheduling problem. In this paper, we study the cascaded flowshop joint scheduling problem, including a distributed permutation flowshop scheduling problem and a hybrid flowshop scheduling problem. The objective is to minimize the makespan, total flowtime, and total tardiness simultaneously. This paper first formulates the working principle and establishes a mixed-integer linear programming model. Second, a two-phase cascaded memetic algorithm (TCMA) is proposed to solve the multi-objective cascaded flowshop joint scheduling problem, whose search space varies sequentially in two distinct production phases to explore as many Pareto-optimal solutions as possible. Combining the problem-specific characteristic, a decomposition-based multi-objective heuristic is proposed to generate a high-quality initial population. Meanwhile, a limited range crossover operator, a limited range mutation operator, and a limited range local search operator are proposed to explore more valuable solution space. The key parameters and operators of the TCMA are calibrated and analyzed using the Taguchi method of design-of-experiment. Computational experiments and comparisons demonstrate the effectiveness of the proposed TCMA for the considered problem.</p>