Dynamic modeling for product family evolution combined with artificial neural network based forecasting model: A study of iPhone evolution

• Published in: Technological forecasting & social change Vol. 178; p. 121549
• Main Authors: Biswas, Sumana, Ali, Ismail, Chakrabortty, Ripon K., Turan, Hasan Hüseyin, Elsawah, Sondoss, Ryan, Michael J.
• Format: Journal Article
• Language: English
• Published: New York Elsevier Inc 01.05.2022; Elsevier Science Ltd
• Subjects: Analytic hierarchy process; Artificial neural networks; Case studies; Dynamic model; Dynamic models; Evolution; Family research; Forecasting; Forecasting model; Fuzzy analytical hierarchy process; Grey relational analysis; Model accuracy; Network reliability; Neural network; Neural networks; Organizational effectiveness; Product family evolution; Reliability analysis; Smartphones; Time dependence; Grey relational analysis; Forecasting model; Neural network; Fuzzy analytical hierarchy process; Dynamic model; Product family evolution
• ISSN: 0040-1625; 1873-5509
• DOI: 10.1016/j.techfore.2022.121549

•The proposed model combines the evolution model with the forecasting model.•It considers market demand, customer needs and technological requirements together.•The ANN model can forecast the specification of some features of future product.•The proposed model is applicable in the engineering design of product development. Products continuously evolve over time. Realizing the pattern of product family evolution along with proper estimation of features for future products has been regarded as a critical issue for business success. Focusing on this issue, a dynamic model for product family evolution combined with forecasting is proposed in this research work. The proposed model considers the influence of market demand, customer needs and technological requirements that are time-dependent. The methodology is a four-phase model. In this model the evaluations of product family evolution are based on the Grey Relational Analysis and Fuzzy Analytical Hierarchy Process. Sensitivity is performed to investigate the reliability of the model. In addition, a data-driven neural network-based forecasting model is proposed that can forecast the specification of the most influential features of future product with a reasonable accuracy. This forecasting model utilizes the information of the previous generation’s product. For each phase, the effectiveness of the developed approach is demonstrated with numerical simulation results and validated with a case study of Apple’s iPhone product family. The case study shows that the approach is very effective to identify the most influential key design features and best performed products that will influence the evolution design of future product.