ABSTRACT
Laser cutting is one of the most important non-conventional manufacturing
processes due to its unique properties and advantages. It is a multi-parameter
technology. This fact has resulted in some gaps in available information, and a few
of contradictory reports about the exact effect of each process parameter on the
surfaces produced, and makes quick parameters selection for obtaining a desired
cut quality problematic, especially for new laser types. Fiber lasers have only very
recently come into prominence, and the literature is still very sparsely populated with
data on their performance. This paper outlines the use of a scoring model of
Projection to Latent Structure (PLS) technique and applies it to study the effects of
fiber laser cutting process parameters and their interactions on the quality achieved
during cutting mild steel sheets. The effects of the interaction of the main process
parameters on the quality output variables such as top kerf width, surface roughness
and the striation depth are studied. Scoring models are proposed to relate the quality
and the laser cutting parameters.