Criticality Evaluation for Spare Parts Based on BP Neural Network

Based on the experience of operations and support, the criticality class of spare parts (SPs) is usually uncertain and may result in excess or insufficient inventory. So it´s an urgent issue to devise a way to evaluate the criticality class of SPs accurately. The investigation applied back-propagation network (BPN) to evaluate the criticality class (i, II, III, IV) of spare parts. By using group-discussing and anonymous questionnaire methods, the index set for the evaluation of criticality class of SPs was put forward. Then the implementation of the evaluation model was depicted in detail. The results show that the model can evaluate the criticality class of SPs effectively, which can avoid the influence of human factors and fuzzy-random city. The proposed BPN will successfully decrease inventory holding costs by modifying the unreasonable target service level setting which is decided by the criticality class and can provide some references for the inventory management.