Studying Evolutionary Solution Adaption by Using a Flexibility Benchmark Based on a Metal Cutting Process.

Area of Science:

• Manufacturing Process Optimization
• Computational Intelligence
• Bio-inspired Algorithms

Background:

• Manufacturing process optimization is a multi-objective challenge with conflicting goals like quality and time.
• Re-optimizing parameters for changing production needs requires costly simulations.
• Reducing the number of required simulations is crucial for efficiency.

Purpose of the Study:

• To develop a bio-inspired framework for system flexibility in optimization tasks.
• To investigate the transferability of solutions across related optimization problems.
• To reduce the computational cost of adapting manufacturing process parameters.

Main Methods:

• Introduced a multi-objective optimization benchmark based on the extended Oxley model for orthogonal metal cutting.
• Studied the flexibility of the Non-dominated Sorting Genetic Algorithm II (NSGA-II).
• Developed two variants of NSGA-II: varying goals and active-inactive genotype.

Main Results:

• Standard NSGA-II significantly reduced the number of evaluations needed for optimization.
• The proposed variants further improved adaptation efficiency, reducing computational effort by over 50% on average.
• The study demonstrated the effectiveness of transferring solutions for dynamic optimization tasks.

Conclusions:

• The developed bio-inspired framework and NSGA-II variants enhance system flexibility and reduce optimization costs.
• These methods show promise for dynamic evolutionary optimization in manufacturing.
• Further research is needed to apply these techniques effectively in real-world industrial applications.