Abstract: Inner-guided hydraulic cylinders are key components of aluminum alloy casting machines, and cranes play an indispensable role in the production. To address this issue, this study establishes a flexible job shop scheduling model integrating processing, transportation, and loading/unloading operations. Furthermore, an improved grey wolf optimizer is proposed to solve the established multi-objective optimization problem for minimizing makespan, equipment energy consumption and crane workload variance. In the decoding stage, crane scheduling rules are explicitly considered. The initial population is generated using the theory of good point sets along with a hybrid strategy to balance diversity and quality. Additionally, an adaptive hunting weight coefficient, a nonlinear convergence factor, and a velocity-assisted term are introduced to enhance the optimization performance, while high-quality solutions are maintained in an external archive. Four neighborhood search structures are further developed to strengthen local search. Comparative experiments using multiple performance indicators show that the improved grey wolf optimizer demonstrates its superior in solving the cooperative scheduling problem considering crane operation constraints.