In recycling and waste recovery, an optical sorter usually combines machine vision with NIR, colour or multisensor detection so the system can recognise polymer type, colour, texture or contamination before separation. AISORT platforms are used for PET, HDPE, PP, PS, PVC, mixed plastics, glass, metals and other recoverable streams.

Use it when manual sorting is too slow, mechanical sorting is not pure enough, or downstream buyers require tighter output consistency.

Mechanical equipment is effective for screening, size control and rough pre-separation. AI sorting is used when your line must distinguish look-alike materials, improve purity, reduce labour dependency or stabilise output over long production hours.

• Mechanical sorting is usually the first-stage separation.
• AI sorting is usually the upgrade layer that increases product value.
• The best results often come from combining both in one line.

AISORT systems are designed for dirty, mixed and variable streams, but recovery performance still depends on upstream preparation such as dosing, singulation, size control and contamination removal. For mixed plastics, the line design matters as much as the sensor itself.

If your stream contains heavy overlap, black plastics or large contamination swings, the project should be assessed as a system problem, not only as a standalone machine purchase.

The S-Series uses NIR spectroscopy with AI vision to identify polymer families such as PET, HDPE, PP, PS and PVC, then separates target fractions with air jets. It is most valuable where the line needs vertical footprint efficiency, high-speed sorting and repeatable output quality.

Related page: AI Tower Optical Sorter S1.

Tower sorters typically operate in the 1 to 8 t/h range per unit, while high-speed vision sorters can reach roughly 3 to 15 t/h in suitable applications. Actual capacity is limited by singulation quality, particle size, burden depth and whether the line prioritises maximum throughput or tighter purity.

For meaningful quotations, throughput should always be evaluated together with feed description and output specification.

Return on investment is usually driven by four variables: labour reduction, recovered material value, reject reduction and downstream buyer acceptance. In stronger projects, operators report 30 to 60 percent labour cost reduction after automation and higher sale value due to better purity consistency.

ROI should be modelled from real plant data rather than catalogue assumptions, especially for retrofit projects.

The exact purity target depends on local regulatory and buyer requirements, but bottle-to-bottle lines are generally much less tolerant of off-spec contamination. That means sorting stages should be designed around the contamination sources that most directly damage downstream reprocessing quality.

Related page: Bottle-to-Bottle Solution.

Project support can include installation guidance, commissioning, parameter adjustment, operator training and post-startup tuning. The exact support model depends on geography, partner network coverage and whether the project is a machine supply or a wider line integration scope.

Lead time depends on machine configuration, sensor package, fabrication queue and whether the order includes wider integration content. Retrofit projects may also add preparation time because electrical, structural and layout checks need to be closed before shipment.

Utility requirements vary by model and line scope, so power, compressed air, footprint and control interfaces should be confirmed against the exact platform before final quotation. For retrofit lines, utility compatibility should be checked early to avoid delay during installation.

AISORT maintains spare parts support and can ship internationally through major courier channels. For operators with uptime-sensitive lines, the more important planning question is which wear parts or critical components should be stocked locally before startup.