A Local Density-Based Abnormal Case Removal Method for Industrial Operational Optimization under the CBR Framework

Operational optimization is essential in modern industry and unsuitable operations will deteriorate the performance of industrial processes. Since measuring error multiple working conditions are inevitable practice, it necessary to reduce their negative impacts on operational under case-based reasoning (CBR) framework. In this paper, a local density-based abnormal case removal method proposed remove cases retrieval step, so as prevent deterioration optimization. More specifically, reasons why classic CBR would retrieve analyzed from perspective industry. Then, algorithm designed based Local Outlier Factor (LOF), properly integrated into traditional step. Finally, effectiveness superiority was tested by numerical simulation an study cut-made process cigarette production. The results show that improved 23.5% compared with CBR, 13.3% fuzzy reasoning.