Hybrid deep learning cost evaluation using CNN with ANN for the plastic injection industry.

The plastic injection moulding industry has been growing and expanding rapidly. Cost evaluation is important to business operations and numerous aspects influence part pricing. However, the intricacy of production data often includes distinctive factors that result in inaccurate and a long waiting time. Therefore, the objective of this research is to propose a cost evaluation approach that combines a 3-dimension convolutional neural network (3D-CNN) with an artificial neural network (ANN) to improve the accuracy of complex geometry from a dense voxel of convolutional neural network (CNN) that can disentangle the difficulty of primary cost evaluation. The methodology consists of 3D-voxelization adopted to 3-dimension convolutional neural network (3D-CNN) and the feature extraction of complex geometry to feature parameters using the learning ability of artificial neural network (ANN) to achieve better accuracy. Then, bulk price analysis is developed for multi-price in-depth for multi-volumes. These results can predict cost evaluation at about 98.65% accuracy for parts costs, 95.17% accuracy for mould costs, and 96.83% of multi-price for multi-volume. The contribution of this research is based upon a new hybrid deep learning using 3-dimension convolutional neural network (3D-CNN) with artificial neural network (ANN) that is practical and accurate in performing cost evaluation multi-prices for multi-volumes for decision-making in the plastic injection industry. [ABSTRACT FROM AUTHOR]