Human-machine fusion-based operational complexity measurement approach to assembly lines for smart manufacturing

Complexity is an important quantification of uncertain operation in assembly lines and the key source of invisible uncertainty problems in smart manufacturing. The purpose of this paper is to propose a complexity measurement approach to assess the complexity of assembly lines integrating humans, machines and configurations. First, the complexity models of the three states of the operation related to humans and machines are built based on information entropy and the operation time model. Then, an operational complexity model is built at the station level; it is constructed with a single station, parallel stations and sublines based on Kolmogorov entropy. The model quantitatively describes the cumulative complexity along with the material flow. Furthermore, the complexity model of the overall system is given, and the Lempel-Ziv algorithm is applied to measure the complexity flow along with the stations. The complexity equilibrium index is derived to quantify the balancing degree among the stations. The model incorporates uncertain operation into system modeling to quantify the influence of uncertainties on the state of the assembly line. An engine assembly line is used to validate that the approach can measure the complexity from operation to station to system.