Abstract

This research underscores the prospect of geospatial analysis in machining operations to enhance precise prediction and robustness, offering a comprehensive framework of spatial modeling for advanced manufacturing processes. Geospatial analysis not only provides accurate predictions but also estimates the uncertainty associated with these predictions, offering valuable insights for process optimization. The surface quality in the machining processes is expressed by the estimation of the average surface roughness. While machining parameters are extensively analyzed for their influence on surface quality, the roughness profile parameters are inadequately explored. This work integrates these underexplored parameters into geospatial predictive models and evaluates their impact on the prediction of average surface roughness in turning operations. The spatial model is validated using a standard method, specifically Leaving-One-Out-Cross-Validation (LOOCV), to ensure reliability and accuracy. The findings of the proposed method demonstrate the potential of the spatial surrogate model in surface quality prediction, particularly in its ability to incorporate spatial dependencies based on roughness profiles while investigating the sensibility of the sample size.

Meeting Name

IISE Annual Conference & Expo 2025

Department(s)

Engineering Management and Systems Engineering

Keywords and Phrases

Machining, Turning, Surface Roughness, Spatial Analysis, Universal Kriging

Document Type

Article - Journal

Document Version

Citation

File Type

text

Language(s)

English

Publication Date

31 May, 2025

Download

Full Text Link

Share

 
COinS