Generating interpretable fuzzy rules for adaptive job dispatching

Abstract

Adaptive scheduling is an approach that selects and applies the most suitable strategy considering the current state of the system. The performance of an adaptive scheduling system relies on the e ectiveness of the mapping knowledge between system states and the best rules in the states. This study proposes a new fuzzy adaptive scheduling method and an automated knowledge acquisition method to acquire and continuously update the required knowledge. In this method, the criteria for scheduling priority are selected to correspond to the performance measures of interest. The decisions are made by rules that re ect those criteria with appropriate weights that are determined according to the system states. A situated rule base for this mapping is built by an automated knowledge acquisition method based on system simulation. D istributed fuzzy sets are used for evaluating the criteria and recognizing the system states. The combined method is distinctive in its similarity to the way human schedulers accumulate and adjust their expertise: qualitatively establishing meaningful criteria and quantitatively optimizing the use of them. As a result, the developed rules may readily be interpreted, adopted and, when necessary, modi ed by human experts. An application of the proposed method to a job-dispatching problem in a hypothetical exible manufacturing system (F M S) shows that the method can develop e ective and robust rules.

A fuzzy dispatching strategy for due-date scheduling of FMSs with information delays

“…Simulation results demonstrate the effectiveness of the proposed system. Lee et al (2001) propose a fuzzy adaptive scheduling method for an FMS part dispatching problem. The method uses an automated knowledge acquisition system to develop effective and robust scheduling rules for the dynamically evolving operational FMS environment.…”

A Fuzzy Nonlinear Programming Approach for Optimizing the Performance of a Four-Objective Fluctuation Smoothing Rule in a Wafer Fabrication Factory

“…Murata et al [2] used six fuzzy rules to move jobs between different priority classes. Lee et al [8] established two fuzzy inference rules to select a combination of some existing dispatching rules for scheduling a flexible manufacturing system. Tan and Tang [9] applied Taguchi's design of experiment (DOE) techniques to improve the design of some fuzzy dispatching rules for a test facility.…”

A Fuzzy Rule for Improving the Performance of Multiobjective Job Dispatching in a Wafer Fabrication Factory

“…System type Rules/inputs/outputs Number of objectives Xiong et al [4] TSK 2/2/1 2 (average cycle time, lateness) Benincasa et al [5] Mamdani 27/3/1 2 (average cycle time, WIP level) Lee et al [9] Single-rule-based 2/4/1 2 (average cycle time, cycle time standard deviation) Tan and Tang [10] Mamdani 8/4/1 3 (throughput, average cycle time, no. of vehicles) Dong and Liu [11] ANFIS 16/4/1 1 (average cycle time)…”