Title: Intelligent optimisation scheduling of raw material area in sausage production line

Authors: Zhonghua Han; Mingpeng Guan; Hangyu Liu; Daliang Chang; Liangliang Sun

Addresses: School of Electrical and Control Engineering, Shenyang Jianzhu University, Shenyang – 110168, Liaoning, China ' School of Electrical and Control Engineering, Shenyang Jianzhu University, Shenyang – 110168, Liaoning, China ' School of Electrical and Control Engineering, Shenyang Jianzhu University, Shenyang – 110168, Liaoning, China ' Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang – 110017, Liaoning, China ' School of Control Engineering, Northeastern University at Qinhuangdao, Qinhuangdao – 066099, Hebei, China

Abstract: The production process in the raw material area of the sausage production line needs to be further optimised to improve production efficiency and reduce the waiting time for suctioning of emulsifying fillings. In this study, we propose an improved Q-learning method for scheduling in raw material area (RMSPS-IQL). To overcome the problem of low learning efficiency caused by multi-operation and multi-tasking in the production line scheduling process, a priority stranding probability is added to the action selection method based on the epsilon-greedy strategy. In addition, to determine the execution time of the action during continuous working, a stranding action starting control method based on the probability of filling deterioration was designed. Comparative analysis of the results of multiple simulation schemes suggested that the RMSPS-IQL effectively reduces the waiting time for suctioning fillings from emulsifying pot, and improves the productivity of the production line.

Keywords: sausage production line; raw material area; intelligent scheduling; probability of filling spoilage; Q-learning.

DOI: 10.1504/IJSPM.2023.138580

International Journal of Simulation and Process Modelling, 2023 Vol.20 No.4, pp.239 - 254

Received: 22 Sep 2023
Accepted: 04 Dec 2023
Published online: 13 May 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article