Cognitive computing based manufacturing data processing for internet of things in job-shop floor

The internet of things (IOT) technology is introduced to the job-shop floor to address the barriers between upper management system and the underlying field automation systems, but also brings new problems, namely, the production management data appears explosive growth. For solving this problem, a data processing methodology based on cognitive computing and cognitive informatics is presented. By simulating the human brain information processing to eye, ears, hands, nose, tongue and other sensory organs, the data of job-shop floor IOT is divided into seven layers from bottom to top, and those layers is classified to passive data acquisition layer and active data acquisition layer. Active data acquisition process consists of three phases. The manufacturing data, manufacturing information and manufacturing knowledge are respectively acquired in three phases, and stored in the corresponding type of database in order to realize the fast reading and updating in different ways. Manufacturing information and manufacturing knowledge based on different granularity are also divided into different levels to meet the job-shop floor IOT management requirements for quick and correct decision-making. This methodology not only effectively reduces the scale of job-shop floor IOT management data, but also gives out different judgments for manufacture problems according to different response time requirements. What´s more, five key enabling technologies are described in detail, that is, layered reference model of data, the database function model, data processing stage division, manufacturing information acquisition model and manufacturing knowledge hierarchical model.