Scheduling Optimization of Flexible Workshop Dynamics Based on Big Data

Modern machining has the characteristics of “small batch, multiple machining types, and complex processes”. The production process generates a large amount and variety of data, and the production strategy is complex and variable. The generation of scheduling strategies needs to consider the actual production status of the workshop and the disturbances in the current status to ensure that the scheduling results meet the actual order requirements. This paper proposes a dynamic scheduling optimization strategy based on workshop big data. We design a big data-based scheduling optimization computing framework, which integrates big data resources such as workshop machine tool processing status, workpiece processing information, and order information to obtain relevant data affecting processing in real-time, and quickly respond to changes in workshop factors to ensure consistency between scheduling constraints and actual situations. In addition, we have improved the traditional genetic algorithm and the method of calculating the fitness value, solving the problem of deviation from the original scheduling plan caused by the disturbance of urgent order insertion, rework and repair, machine tool maintenance, and scheduling time. This improves the real-time and dynamic decision-making of machining scheduling. The proposed scheduling optimization strategy has been applied to a certain aviation engine machining workshop, reducing the overall machining time and improving machining efficiency compared to the original scheduling strategy of the workshop.