Design and Numerical Assessment of an Automated, Sensor-Supervised PET Bottle Shredding System for Small-Scale Recycling

Abstract

This study presents the design and numerical assessment of an automated, sensor-supervised PET bottle shredding system for small-scale recycling, where manual feeding exposes the operator to the blade assembly and delivers material unevenly. An inclined cleated conveyor is coupled to a single-shaft shearing mechanism under near-infrared, Hall-effect and motor-current supervision, with a stop-reverse-retry jam-protection sequence. Requirements from a structured survey (n = 50) were mapped to technical specifications by quality function deployment, and materials and processes were selected by weighted digital logic. Six components were evaluated by static structural finite element analysis and cross-checked against independent closed-form calculations at the as-built geometry. The cross-check located a factor-of-ten torque error in the source shaft model; correcting it raises the shaft equivalent stress from 20.91 to 23.0 MPa, a 10% change, since bending governs at the Ø65 mm journal. Against the yield strengths of the specified materials, the minimum static factor of safety across the assembly is approximately 8 at the rotating cutting blade, with the shaft at 13.5 and a modified-Goodman fatigue factor of safety of 6.5. Mounting the blades on an 80 mm across-flats hexagonal section removes the keyed blade interface that governs comparable single-shaft designs. A five-year capacity ramp model gives a base-case net present value of BDT 36.2 crore, positive in all 10,000 Monte Carlo iterations, and per-unit processing is estimated at 140 t PET yr⁻¹, a net avoidance potential of 236 t CO₂e yr⁻¹. The work provides a design-stage basis and ranked verification programme for prototype construction.