| dc.description.abstract | This thesis proposes concept of smart manufacturing is combined with the parameters required by the microfluidic process method, and at the same time, the manufacturing yield, cost and time are considered to achieve the process-based identification method to achieve the purpose of microfluidic commercialization.
According to the front-end material resolution , roughness , cost , and output , we set 25μm , 50μm , 40K , 5min and the back end bonding strength , and the bonding time set 20mpa , 5min to distinguish the module discriminant flow direction , and the process will be based on the input ideal The condition value of the flow to the final manufacturing method , and the actual case is extracted to actually verify the correctness of the predicted flow direction .Finally ,the predicted result method is converted into a valid ID value and the production lines production process of the factory is actually carried out in the CIROS software . This factory can According to the trays are sequentially transported to the processing station for signal identification, processing is performed through the signal identification set by the machine, and each processing station can complete the work autonomously, and the operation of the robotic arm can accurately locate and grasp the material and position. Practicing the proof of concept of a small number of diverse products of the smart factory, the efficiency of the factory can be significantly improved by replacing the manual operation, and the manpower in the process only needs to be supervised to monitor the product quality. | en_US |