Using Artificial Intelligence to develop a prediction model, increasing the viability of previously unrecyclable, low grade, post-consumer polymers; promoting Clean Growth through business efficiency and helping to achieve Net Zero - ColourAI

塑料回收商面临的一个关键挑战和生产回收聚合物的一个主要限制因素是基础聚合物投入材料的差异很大，这意味着满足客户的规格非常具有挑战性，每一批都需要独特的配方，必须通过实验获得；不同于原始聚合物混炼机-它有一个标准配方，可以始终如一地产生具有所需性能的化合物。由于混合颜色之间的关系不是线性的(例如，蓝色加黄色输入并不总是产生绿色)，因此执行多次试错迭代以匹配颜色规范。配制批次可能平均需要5次迭代才能获得所需的颜色--在某些情况下，这可能需要20次迭代。这项第一阶段的技术可行性研究及后续的第二阶段原型系统开发及测试计划，将有助推行政府的清洁增长策略，让塑胶回收商及复合商可因生产多个测试批次的循环再造物料以达到客户的颜色规格而降低加工成本，从而提高商业及工业效率。

A key challenge for plastic recyclers and a major limiting factor in producing recycled polymers, is the wide variation in base polymer input materials which means that meeting the customer specification is very challenging and every batch requires a unique formulation that must be derived experimentally; unlike virgin polymer compounders - who have a standard formulation which will consistently yield a compound with the desired properties. Multiple trial and error iterations are performed to match the specification for colour, as the relationship between mixed colours is not linear (e.g. blue plus yellow inputs do not always produce green). Formulating batches can take an average of 5 iterations to achieve the desired colour - and in some instances this can take up to 20 iterations. This phase 1 technical feasibility study and follow-on phase 2 prototype system development and testing project will help deliver the Government's Clean Growth Strategy by enabling plastic recyclers and compounders to improve business and industry efficiency by reducing processing costs experienced by having to produce multiple test batches of recycled material in order to achieve the customer's colour specification.