Optimal-Parameter Determination by Inverse Model Based on MANFIS: The Case of Injection Molding for PBGA

This paper presents a novel method of integrating both optimization and inversely modeling methods to determine the optimal input-parameter for a multi-input multi-output (MIMO) system to realize the desired output-performance. First, the Taguchi method is employed to minimize experimental numbers and to collect experimental data representing the quality performances of a MIMO system. Next, the MANFIS is used to train the inverse model based on the data from the Taguchi experimental method. The adaptive neuro-fuzzy inference system (ANFIS) has been widely used for modeling different kinds of nonlinear systems. In this study, the method is further extended to MIMO-ANFIS (MANFIS) architecture to train the inverse model. The well-trained model has the ability of uniquely determining the inverse relationship for each input-output set. A case study involving statistical characterization and multiple criteria optimization on injection molding for plastic ball grid array (PBGA) is successfully presented to demonstrate the effectiveness of the proposed method.