Hybrid Simulation Modelling to Support the Development of Recycling Infrastructures
An enormous amount of waste material arises internationally due to linear economic structures in which natural materials are converted into products which are discarded after use. This problem could be addressed by a shift towards a circular economy, whereby products are kept in use for longer and the material is recovered at end-of-life as a valuable resource. Recycling is therefore central to the development of a circular economy. However, recycling levels are low in many countries, often due to lack of recycling infrastructures. This research aims to support the development of waste management and recycling infrastructures by developing a model to experiment with new systems, particularly looking at the balance between centralised and delocalised systems for collection, sorting and recycling.
The model utilises hybrid simulation modelling due to the accuracy and flexibility that the method provides. The model was developed using an agent-based architecture which includes three submodels to describe the process, logistics, and communication. With this architecture, the model has the capability to simulate a wide range of waste management systems. The frameworks developed in this research provide a new approach to using hybrid simulation modelling. Moreover, the frameworks can be used as guidelines for the development of future tools and models.
The model was implemented into two case studies, in different geographic locations. The first case study investigated the waste management structures in Singapore and compared the performance of centralised and distributed systems. The results showed that the facility processes are more favourable in a centralised system due to economies of scale. On the other hand, the logistics perform better in a distributed system. The balance between centralised and distributed structures can be analysed using the model developed in this research, informing decision-making processes for future developments. The second case study applied the model to analyse the potential for introducing new waste management infrastructures in Indonesia, working with a company currently tasked with reducing waste mismanagement in small cities and rural areas. The results validated the model capability in experimenting with new systems and offered insights into the relative performance of centralised and distributed waste management systems. The work has provided a valuable additional dimension to the feasibility studies and contributes to the recommendations. Both case studies have demonstrated the viability and applicability of this hybrid simulation model as a tool that can be adapted to optimise waste management infrastructures in different environments.