Product platform design and customization: Status and promise
Artificial Intelligence for Engineering Design, Analysis and Manufacturing - SPECIAL ISSUE: Platform product development for mass customization
A fast and elitist multiobjective genetic algorithm: NSGA-II
IEEE Transactions on Evolutionary Computation
A mixed-discrete Particle Swarm Optimization algorithm with explicit diversity-preservation
Structural and Multidisciplinary Optimization
| Hi-index | 0.00 |
A product family with a platform paradigm is expected to increase the flexibility of the manufacturing process to market changes, and to take away market share from competitors that develop one product at a time. The recently developed Comprehensive Product Platform Planning (CP3) method (i) presents a generalized model, (ii) allows the formation of product sub-families, and (iii) provides simultaneous identification and quantification of platform/scaling design variables. The CP3 model is founded on a generalized commonality matrix that represents the platform planning process as a mixed-binary nonlinear programming (MBNLP) problem. This MBNLP problem is high-dimensional and multimodal, owing to the commonality constraint. In this paper, the complex MBNLP model is reduced to a tractable MINLP problem without resorting to limiting approximations; along the reduction process, redundancies in the original commonality matrix are also favorably addressed. To promote a better understanding of the importance of a reduced MINLP, this paper also provides a uniquely comprehensive formulation of the number of possible platform combinations (or commonality combinations). In addition, a new commonality index (CI) is developed to simultaneously account for the inter-product commonalities (based on design variable sharing) and the overlap between groups of products sharing different platform variables. To maximize the performance of the product family and the commonality index yielded by the new CP3 model, we apply an advanced mixed-discrete Particle Swarm Optimization algorithm. The potential of the new CP3 framework is illustrated through its application to design scalable families of electric motors. Maximizing the new CI produced families with more commonality among similar sets of motor variants (compared to maximizing the conventional CI), which can be a beneficial platform attribute for a wide range of product families.