Reducing packaging waste requires organizations to look beyond their own products, services, and business models. Collaboration in circular ecosystems may offer a promising approach. Material flows in circular ecosystems are affected by social, economic, and technical variables,
...
Reducing packaging waste requires organizations to look beyond their own products, services, and business models. Collaboration in circular ecosystems may offer a promising approach. Material flows in circular ecosystems are affected by social, economic, and technical variables, including decision-making behaviour, material prices, and available technologies. The complexity of these interactions makes it challenging to assess the impact of strategic choices on circular ecosystems' effectiveness in reducing waste. Agent-based modelling (ABM) is a useful methodology for analysing dynamics within such complex systems. This study develops an ABM for a Dutch food packaging ecosystem and integrates organizational decision-making theory to account for actor behaviour, considering different decision-styles and rules. The ABM includes three types of agents representing beverage producers, packaging producers, and waste treaters, who can form circular ecosystems for closed-loop recycling. Experimentation indicates that with just 10 % of organizations prioritizing circularity over maximizing individual profit, significant waste reduction is achievable, although the decision-style of the beverage producer is crucial in profit-driven ecosystems. Furthermore, centralized waste management could stabilize recycled material supply and mitigate fluctuations in recycled content. While the model is limited by deterministic agent behaviour and simplified decision-making processes, our findings demonstrate the value of ABM in understanding dynamics in circular ecosystems and provide insights for policymakers and industry stakeholders. Future research could explore alternative circular strategies with ABM, such as packaging reuse and material substitution, and the impact of modelling more nuanced decision-making behaviour on the model's scientific and practical value.