Kharazmi University International Journal of Supply and Operations Management 2383-1359 13 2 2026 05 01 Optimization of Key Parameters in a Welding Process Using Multivariate Statistical Analysis and Design of Experiments 244 272 11087 10.22034/ijsom.2026.110844.3429 EN Narges Delavar Pasikhani Department of Industrial Engineering, Sharif university of technology, Tehran, Iran Abbas Foroozanfar Department of Industrial Engineering, Sharif university of technology, Tehran, Iran Seyed Taghi Akhavan Niaki Department of Industrial Engineering, Sharif university of technology, Tehran, Iran Journal Article 2025 07 08 Objective: Welding quality strongly affects product performance, durability, scrap rate, and production cost. This study aims to develop an integrated data-driven framework combining Principal Component Analysis (PCA), Linear Discriminant Analysis (LDA), and Design of Experiments (DOE) to optimize welding parameters and improve weld quality in an industrial manufacturing process.Methods: Historical production data from a turbine manufacturing company in Iran were analyzed. The investigated welding parameters included current intensity, welding position, welding speed, preheat temperature, and electrode diameter. A 2k factorial design was applied to evaluate their effects on weld depth, which was maximized, and weld width, which was minimized. Multi-response optimization was conducted using response surface methodology and desirability functions, with different weighting schemes assigned to the response variables. Results: The findings revealed that current intensity was the most influential parameter affecting weld quality. Moreover, a significant interaction effect was identified between current intensity and welding position, indicating that the joint effect of these parameters plays an important role in determining weld profile characteristics. Compared with the historical data, the optimized parameter settings improved weld depth by an average of 13.77% and enhanced weld width by an average of 3.56%. Following the implementation of the optimal parameter values in the production process, the average scrap rate decreased from 19.8% to 13.1% over a six-month period. Conclusion: The proposed framework effectively improved weld quality and production performance. The reduction in scrap rate and enhancement of welding responses resulted in an estimated 3–4% decrease in production costs, demonstrating the practical value of the approach for industrial welding optimization. http://www.ijsom.com/article_11087_a82aac2169c9a76e374b2f4c96bfa0a0.pdf