Simulated annealing: theory and applications
Simulated annealing: theory and applications
Genetic algorithms + data structures = evolution programs (2nd, extended ed.)
Genetic algorithms + data structures = evolution programs (2nd, extended ed.)
Feature-based cost estimation for packaging products using neural networks
Computers in Industry
Genetic Algorithms in Search, Optimization and Machine Learning
Genetic Algorithms in Search, Optimization and Machine Learning
PKDD '97 Proceedings of the First European Symposium on Principles of Data Mining and Knowledge Discovery
Genetic Operators in a Dual Genetic Algorithm
TAI '95 Proceedings of the Seventh International Conference on Tools with Artificial Intelligence
Intelligent FEA-based design improvement
Engineering Applications of Artificial Intelligence
International Journal of High Performance Computing Applications
Hi-index | 0.00 |
Manufactured products are commonly encased in moulded protective packaging buffers to protect them from damage due to impact shock during handling and transportation. The materials used to fabricate these buffers are well-known and cost-effective but not friendly to the environment. New bio-degradable materials such as paper pulp and starch have emerged as formidable alternatives, but little is known about how to design buffers from them.This paper describes a novel intelligent methodology for the virtual modeling, testing and design of protective packaging buffers. The methodology employs the use of genetic algorithms, finite element model and design routines developed to determine the optimal buffer design. Based on an ANSYSTM finite element model of the buffer, simulated drop tests were performed. The magnitudes of the largest reaction forces for the simulated drop tests as encountered by the model are computed and translated into the highest G value that the buffer can sustain without damage to the product. From the results, a more superior set of buffer designs is then derived with each passing generation.Validation tests were conducted on six different buffer configurations designed to protect six common consumer electrical appliances. The simulated G values were found to differ by a maximum of 11.8% from empirical results. The industrial norm of 10% deviation between empirical and simulated values can easily be realized when further refinements are made to the basic finite element model of the buffer. The findings validate the new methodology in buffer design in particular for new packaging materials where there are only a limited number of explicit or heuristic design rules.