Mixed modeling approach for mechanical sorting processes based on physical properties of municipal solid waste

Material recovery facilities (MFRs) play an important role in today's waste management systems to maximize recycling efficiency for several waste materials. These facilities face multiple challenges, often due to a poor understanding of the mechanisms occurring within the sorting equipment. Improving modeling techniques of these unit operations appear to be a promising opportunity to mitigates these challenges. Mechanical sorting efficiency of municipal solid waste is often predicted from simple transfer coefficients, which are obtained for a specific set of operating conditions of the sorting equipment and sorting sequence configuration. When these transfer coefficients are used in situations that are different to those in which they were obtained, poor predictions can be expected. To overcome these limitations, a new predictive tool, based on the integration of mechanistic models and transfer coefficients, is presented. Mechanistic models are developed only for the most influential unit operations in a MRF, in order to predict their sorting efficiency based on the physical phenomena occurring. Integration of these models with the use of transfer coefficients for the other unit operations allows the entire predictive tool to remain as simple as possible while providing high prediction accuracy and flexibility. The use of the tool is validated with a real case study of a material recovery facility. Results indicate a good prediction of the mass flows of the facility. Moreover, a new modeling technique is proposed for the representation of a ballistic separator based on the shape factor of the waste items.