Optimization of precision machine part manufacturing by integration of Grey-Taguchi method with principal component analysis

Erol K., Kapan Ulusoy S., Şenyiğit E.

SIGMA JOURNAL OF ENGINEERING AND NATURAL SCIENCES, cilt.44, sa.1, ss.292-308, 2026 (ESCI, Scopus)

Özet

Determining and optimizing the process parameters impacting the outputs at each produc-

tion stage is necessary to reduce production costs. The Taguchi Method (TM) and the Grey

Relational Analysis (GRA) are commonly utilized two techniques for process parameter op-

timization. In precision machine part manufacturing, Computer Numerical Control (CNC)

production is the most critical process. In this study, the objective is to optimize CNC manu-

facturing parameters using TM, GRA and Principal Component Analysis (PCA) in metal sec-

tor. Process parameters like operator experience level (in years), CNC machine brand, CNC

machine age, and CNC machine size were determined and optimized based on their degree

of impact on the outputs. The experiments were carried out using a four-factor, four-level

Taguchi orthogonal array (L16), and Analysis of Variance (ANOVA) was conducted aiming to

determine the effects of these process parameters on production time, dimension conformity,

and surface roughness performance factors. Selection of these input parameters and perfor-

mance factors in the study is to provide a solution to a problem in the company from which

the data are obtained with scientific methods and to contribute to the literature. Utilizing TM,

the optimal values of process parameters are determined as ten years for operator experience,

as Mazak for CNC machine brand, as two years for machine age, and as 500x550x550 for

machine size. Utilizing the combination of GRA and PCA optimal parameter values are deter-

mined as ten years for operator experience, as Yuntes for CNC machine brand, as two years for

machine age, and as 700x450x500 for machine size. A sensitivity analysis was performed using

21 different weight sets for performance factors (production time, dimension conformity, and

surface roughness). Compared to the initial CNC production process parameters, 45%, 95%,

and 504% improvements were obtained in production time, dimension conformity, and sur-

face roughness process parameters. Companies, especially operating in the metal sector, can

benefit from managerial practices by considering the ranking of parameters affecting CNC

production according to the results obtained from this study.