The Combined Multi-objective Optimization Design for a Light Guide Rod

The light guide rod (LGR) has been popularly used for the vehicles, and the automobile lamp industries need mass production to match this trend. This paper aims to develop a systemic way to find the best parameters' combination for the LGR, and the parameters are usually restricted to some levels and random values. In this paper, the LGR example with two optical performances of illuminance flux and uniformity is to be optimized by use of the real-coded genetic algorithm (RGA) and grey relational analysis (GRA). The illuminance flux and uniformity of the best parameters' combination are obtained and compared with the initial set. Comparisons with Taguchi-Grey can improve 5% of gain and comparisons with Pareto genetic algorithm (PaGA) can improve 1.7% of gain. The combined multi-objective optimization can saving 7% time and it is found that the new proposed method has positive gains in performances.