COST-DRIVEN PRODUCT PLATFORM DESIGN OPTIMIZATION AND METAMODELING WITH APPLICATION ON A CASE STUDY FOR A FAMILY OF GUIDING BUSHES

Document Type : Original Article

Authors

1 Basic and Applied Sciences department, faculty of Engineering and Technology, Arab Academy for Science, Technology and Maritime Transport.

2 Design and Production Engineering department, Ain Shams University, EGYPT.

Abstract

In today’s dynamic manufacturing environment, the implementation of product platforms has become increasingly vital to efficiently accommodate late differentiation in product families. With rising product variety and shifting customer demands, manufacturers must adopt strategies that balance cost-effectiveness with flexibility. Additive and subtractive manufacturing technologies have emerged as key enablers of this adaptability, supporting modular product architectures and platform-based designs. Product platform design is influenced by several critical factors, including the fluctuating demand for product variants, the degree of feature commonality among variants, the cost of constructing the base platform, and the additive and subtractive costs incurred during customization. Additionally, precedence relationships in assembly operations significantly affect the platform structure and its economic feasibility.
 
This study investigates the combined effects of these parameters on optimal platform design. A genetic algorithm was used to minimize the total manufacturing cost while considering varying demand scenarios and cost ratios. The analysis led to the formulation of a general metamodel that can directly determine the optimal platform configuration, incorporating feature dependencies and precedence constraints. The model's accuracy and practicality were validated by applying it to a benchmark case study of a family of guiding bushes, yielding platform designs consistent with optimal results.
 
The findings demonstrate that the proposed metamodel provides a practical and computationally efficient alternative to traditional optimization techniques, making it well-suited for real-world applications in platform-based design and variant manufacturing.

Keywords

Files

Share

How to cite

Statistics