The production process of the process industry is generally a continuous or batch process, with the feature of complexity, uncertainty, multi-objective, multi-constraint, multi-resource coordination and so on. With the advent of industry 4.0 and the era of big data, the process industry began to be developed in the direction of digital automation, at the same time, the idea of big data brings us a new direction to solve the process scheduling problem. It has a huge impact on the results of the optimization that the way of organizing and using these data which comes from production process. In this paper, the industrial process of producing glass fiber by alkali-free kiln process has been chosen as an example, the characteristics and shortcomings of the task - source network model are analyzed and a complex network model of process industry scheduling based on big data has been introduced, which accurately links the data in the production process.