Real-time shop floor scheduling method based on virtual queue adaptive control: Algorithm and experimental results

Abstract

Shop floor scheduling requires consideration of the dynamic, time-varying and unpredictable nature of the manufacturing environment. A shop floor adaptive scheduling method is proposed given an abnormal event represented by machine breakdown. Because of the difficulty of predicting the duration of a machine breakdown when it occurs, an adaptive scheduling method based on queue control is proposed that not only guarantees production execution according to the initial schedule but also ensures the operational reliability of the shop floor when a disturbance occurs. A mechanism triggered by a threshold value is introduced to realize the adaptive rescheduling of the shop floor, which ensures real-time and stable shop floor operations. The proposed scheduling/rescheduling algorithm is based on Petri net and ant colony optimization (PN-ACO). The average relative error in the optimal solution for benchmark instances problems from BC data and BR date is 22.32 and 14.61 respectively, which illustrates the feasibility and efficiency of the proposed PN-ACO algorithm to solve flexible job shop scheduling problems (FJSPs).The experimental results confirm that the proposed adaptive scheduling method is effective for solving the dynamic flexible job shop scheduling problem with uncertain repair time after occurrence of machine breakdown.