Assessing the eco-effectiveness of a solid waste management plan using agent-based modelling

Vitor Miranda de Souza*, Jacqueline Bloemhof, Milton Borsato

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review


Municipal Solid Waste Management is yet to be eco-effectively performed, especially in developing countries. In Brazil, a considerable fraction of waste has been improperly landfilled, generating environmental, social and economic problems. In 2018, the government of the state of Paraná released a revised version of its waste management plan, defining improvement strategies to be gradually implemented until 2038. However, these strategies’ eco-effectiveness has not been forecasted, nor the plan was deployed to the regional level. This research aims to fill this gap, downscaling the plan to the region of Norte Pioneiro, simulating its implementation and monitoring environmental and economic benefits. The dynamics of waste generation, collection and disposal are investigated using an agent-based model, considering the four population growth scenarios addressed in the plan. Targets for strategies of waste reduction, collection, source-separation and charging of waste fees are modelled. Multiple simulation runs were performed and outputs assessed and discussed. Results show that, if the plan is thoroughly implemented since 2020, at least 650 kilotons of avoided CO2eq emissions and US$ 40 million in avoided expenditures can be achieved in the most conservative scenario by 2038. Implications from the strategies proposed in the plan are highlighted, and recommendations to improve the plan's eco-effectiveness are outlined.

Original languageEnglish
Pages (from-to)235-248
Number of pages14
JournalWaste Management
Publication statusPublished - 15 Apr 2021


  • Municipal solid waste management
  • net GHG
  • Scenario-based simulation
  • Waste generation forecasting

Fingerprint Dive into the research topics of 'Assessing the eco-effectiveness of a solid waste management plan using agent-based modelling'. Together they form a unique fingerprint.

Cite this