Machine learning driven analysis and development of LD sludge-polypropylene composites for industrial tribo-components

Purpose This study aims to investigate the effect of Linz–Donawitz (LD) sludge reinforcement on the microhardness and abrasive wear behavior of polypropylene (PP) composites, and to optimize and predict the specific wear rate (SWR) using Taguchi design and an artificial neural network (ANN) model for improved performance and sustainable utilization of industrial waste. Design/methodology/approach PP composites reinforced with 0–20 Wt.% LD sludge were fabricated by compression molding. Microhardness was measured using a micro-indentation test. Three-body abrasive wear tests were conducted according to ASTM standards and analyzed using Taguchi L25 design. An ANN model was developed to predict the SWR. Findings The incorporation of LD sludge significantly enhanced the microhardness and abrasive wear resistance of PP composites. Microhardness increased from 0.059 GPa for neat PP to 0.639 GPa at 20 Wt.% filler content. SWR decreased considerably with increasing LD sludge, indicating improved abrasion resistance. Taguchi analysis identified sliding velocity and filler content as the most influential parameters affecting wear behavior. The ANN model accurately predicted SWR with errors below 9%, confirming its effectiveness for reliable wear performance prediction. Originality/value This study demonstrates the effective utilization of industrial LD sludge as a reinforcement in PP composites to enhance microhardness and abrasive wear resistance. It uniquely integrates Taguchi optimization with ANN-based prediction to analyze and forecast wear behavior, highlighting a sustainable pathway for converting steel industry waste into high-performance engineering materials. Peer review The peer review history for this article is available at: https://publons.com/publon/10.1108/ILT-03-2026-0092