US 12,254,426 B2
Production line operation forecast method and production line operation forecast system
Tsan-Cheng Su, Tainan (TW); Hao-Jhe Huang, New Taipei (TW); and Chung-Wei Lin, Hsinchu County (TW)
Assigned to Industrial Technology Research Institute, Hsinchu (TW)
Filed by Industrial Technology Research Institute, Hsinchu (TW)
Filed on Jul. 27, 2022, as Appl. No. 17/874,302.
Prior Publication US 2023/0342675 A1, Oct. 26, 2023
Int. Cl. G06Q 10/04 (2023.01); G06Q 10/06 (2023.01); G06Q 10/0639 (2023.01); G06Q 10/067 (2023.01); G06Q 50/04 (2012.01)
CPC G06Q 10/04 (2013.01) [G06Q 10/06393 (2013.01); G06Q 10/067 (2013.01); G06Q 50/04 (2013.01)] 16 Claims
OG exemplary drawing
 
1. A production line operation forecast method, comprising:
obtaining an online production line work-in-process map at a time point by a processor, generating a plurality of candidate simulated dispatch decisions based on the online production line work-in-process map, and inferring a plurality of production-line work-in-process map changes of the plurality of candidate simulated dispatch decisions at a next time point;
inputting the plurality of production-line work-in-process map changes to a forecast model by the processor, such that the forecast model outputs a plurality of simulated production line operation health indicators of the plurality of candidate simulated dispatch decisions at the next time point;
selecting one of the plurality of candidate simulated dispatch decisions as a scheduling dispatch decision by the processor;
obtaining a plurality of historical production line work-in-process maps by the processor, and generating a plurality of historical production line operation health indicators according to the plurality of historical production line work-in-process maps; and
training the forecast model by the processor according to the plurality of historical production line work-in-process maps and the plurality of historical production line operation health indicators,
wherein the forecast model is a regression model.