An ANFIS-based dispatching rule for complex fuzzy job shop scheduling problem
Mingyu Dong, Min Liu · International Conference on Information Science and Technology · 2011
Complex job shop scheduling problems are mostly NP-hard. When some knowledge is imprecise, e.g., the processing times are denoted by fuzzy numbers, the fuzzy scheduling problems need new methods to be handled. ANFIS has some characteristics of self- learning, the nonlinear mapping and the form of if-then fuzzy rules. So this paper adopts ANFIS to combine the heuristic rules nonlinearly and takes the ANFIS as an adaptive scheduling rule to sort jobs. This is the first attempt to apply ANFIS in fuzzy job shop scheduling problems with parallel machines. The simulation tests show the feasibility of this method and it performs better than the heuristic rules.