Parameter Optimization for Robotic Milling of Ultra‐High Molecular Weight Polyethylene: A Digitally Prepared Central Composite Design‐Response Surface Methodology Framework With Experimental Validation

ABSTRACT

The integration of six‐axis industrial robots in high‐precision manufacturing has expanded rapidly, yet machining advanced thermoplastics such as Ultra‐High Molecular Weight Polyethylene (UHMWPE) remains challenging. Limited structural stiffness in robotic platforms, combined with the viscoelastic behavior of UHMWPE, can degrade surface integrity and dimensional accuracy (final slot width relative to the nominal target). This study presents a digitally prepared and experimentally validated robotic machining–optimization framework for UHMWPE. SprutCAM is used for preexperimental kinematic verification and toolpath planning, and a CCD‐based response surface methodology (RSM) is applied to develop validated second‐order predictive models and a process window for parameter planning. Spindle speed, feed rate, and depth of cut were evaluated for their effects on surface roughness (Ra) and slot width deviation using a KUKA KR120 R2700 robot with a 10 mm flat end mill under a fixed, stiffness‐favorable posture. Significant second‐order models were obtained ( R ²  = 0.9843 for Ra and 0.9258 for dimensional accuracy). Spindle speed was the dominant factor ( p  < 0.0001), and higher rotational speeds reduced Ra to approximately 0.9 µm. Postprocess surface temperatures remained within 32.8°C–38.1°C, well below the UHMWPE heat deflection temperature (79°C), indicating no thermal damage within the investigated domain. Model validation errors ranged from 0.38% to 3.54%. Overall, the framework enables controlled surface quality and predictable dimensional behavior within the investigated parameter ranges and platform configuration, supporting more efficient setup for thermoplastic finishing tasks in flexible robotic cells.