Compounding Process Optimization for Recycled Materials Using Machine Learning Algorithms

Using environmentally friendly industrial practises is one way to lower CO2 emissions and the use of natural resources. There has been a lot of work done in the last ten years to shift the economy from linear to circular. This change requires the replacement of existing raw material acquisition processes with remanufacturing techniques and technologies that are incorporated into current industrial operations. Being forced to transition from linear to circular processes makes this change challenging for enterprises. For this shift to happen faster, we must use the technologies of the fourth industrial revolution. To improve the mixed design of the RAC for various compressive strength classes, gradient boosting in regression trees and the particle swarm optimisation technique were both used. With regard to a wide range of objective compressive strength classes, the hybrid method generated RAC combination designs with a diminished environmental effect and significant cost savings. With the highest proportion of recycled aggregate, at the lowest cost, and with the least negative effects on the environment, the process may also be used to produce environmentally friendly concrete.