Reducing packaging waste is a critical challenge that requires organizations to collaborate within circular ecosystems, considering social, economic, and technical variables like decision-making behavior, material prices, and available technologies. Agent-Based Modeling (ABM) offers a valuable methodology for understanding these complex dynamics. In our research, we have developed an ABM to explore circular ecosystems’ potential in reducing packaging waste, using a case study of the Dutch food packaging ecosystem. The model incorporates three types of agents—beverage producers, packaging producers, and waste treaters—who can form closed-loop recycling systems.

Beverage Producer Agents: These agents represent the beverage company divided into five types based on packaging formats: cans, PET bottles, glass bottles, cartons, and bag-in-boxes. Each producer has specific packaging demands based on product volume, type, weight, and reuse potential. They select packaging suppliers annually, guided by deterministic decision styles: bargaining (seeking the lowest price) or problem-solving (prioritizing high recycled content).

Packaging Producer Agents: These agents are responsible for creating packaging using either recycled or virgin materials. The model assumes a mix of monopolistic and competitive market situations, with agents calculating annual material needs. Decision styles influence their choices: bargaining agents compare recycled and virgin material costs, while problem-solving agents prioritize maximum recycled content. They calculate recycled content in packaging and set prices accordingly, ensuring all produced packaging is sold within or outside the model.

…

Hierarchical problem-solving model
The model simulates a hierarchical problem-solving process in which a manager delegates parts of a problem to specialists, who attempt to solve specific aspects based on their unique skills. The goal is to examine how effectively the hierarchical structure works in solving the problem, the total cost of the process, and the resulting solution quality.

Problem-solving random network model
The model simulates a network of agents (generalists) who collaboratively solve a fixed problem by iterating over it and using their individual skills to reduce the problem’s complexity. The goal is to study the dynamics of the problem-solving process, including agent interactions, work cycles, total cost, and solution quality.

This model was designed to study resilience in organizations. Inspired by ethnographic work, it follows the simple goal to understand whether team structure affects the way in which tasks are performed. In so doing, it compares the ‘hybrid’ data-inspired structure with three more traditional structures (i.e. hierarchy, flexible/relaxed hierarchy, and anarchy/disorganization).

According to the philosopher of science K. Popper “All life is problem solving”. Genetic algorithms aim to leverage Darwinian selection, a fundamental mechanism of biological evolution, so as to tackle various engineering challenges.
Flibs’NFarol is an Agent Based Model that embodies a genetic algorithm applied to the inherently ill-defined “El Farol Bar” problem. Within this context, a group of agents operates under bounded rationality conditions, giving rise to processes of self-organization involving, in the first place, efficiency in the exploitation of available resources. Over time, the attention of scholars has shifted to equity in resource distribution, as well. Nowadays, the problem is recognized as paradigmatic within studies of complex evolutionary systems.
Flibs’NFarol provides a platform to explore and evaluate factors influencing self-organized efficiency and fairness. The model represents agents as finite automata, known as “flibs,” and offers flexibility in modifying the number of internal flibs states, which directly affects their behaviour patterns and, ultimately, the diversity within populations and the complexity of the system.

Agent-based models of organizational search have long investigated how exploitative and exploratory behaviors shape and affect performance on complex landscapes. To explore this further, we build a series of models where agents have different levels of expertise and cognitive capabilities, so they must rely on each other’s knowledge to navigate the landscape. Model A investigates performance results for efficient and inefficient networks. Building on Model B, it adds individual-level cognitive diversity and interaction based on knowledge similarity. Model C then explores the performance implications of coordination spaces. Results show that totally connected networks outperform both hierarchical and clustered network structures when there are clear signals to detect neighbor performance. However, this pattern is reversed when agents must rely on experiential search and follow a path-dependent exploration pattern.

This ABM simulates problem solving agents as they work on a set of tasks. Each agent has a trait vector describing their skills. Two agents might form a collaboration if their traits are similar enough. Tasks are defined by a component vector. Agents work on tasks by decreasing tasks’ component vectors towards zero.

The simulation generates agents with given intrapersonal functional diversity (IFD), and dominant function diversity (DFD), and a set of random tasks and evaluates how agents’ traits influence their level of communication and the performance of a team of agents.

Modeling results highlight the importance of the distributions of agents’ properties forming a team, and suggests that for a thorough description of management teams, not only diversity measures based on individual agents, but an aggregate measure is also required.

…

This model aims to examine how different levels of communication noise and superiority bias affect team performance when solving problems collectively. We used a networked agent-based model of collective problem solving in which agents explore the NK landscape for a better solution and communicate with each other regarding their current solutions. We compared the team performance in solving problems collectively at different levels of self-superiority bias when facing simple and complex problems. Additionally, we addressed the effect of different levels of communication noise on the team’s outcome

This code is for an agent-based model of collective problem solving in which agents with different behavior strategies, explore the NK landscape while they communicate with their peers agents. This model is based on the famous work of Lazer, D., & Friedman, A. (2007), The network structure of exploration and exploitation.

This model builds on inquisitiveness as a key individual disposition to expand the bounds of their rationality. It represents a system where teams are formed around problems and inquisitive agents integrate competencies to find ‘emergent’ solutions.

A series of studies show the applicability of the NK model in the crowdsourcing research, but it also exposes a problem that the application of the NK model is not tightly integrated with crowdsourcing process, which leads to lack of a basic crowdsourcing simulation model. Accordingly, by introducing interaction relationship among task decisions to define three tasks of different structure: local task, small-world task and random task, and introducing bounded rationality and its two dimensions are taken into account: bounded rationality level that used to distinguish industry types and bounded rationality bias that used to differentiate professional users and ordinary users, an agent-based model that simulates the problem-solving process of tournament-based crowdsourcing is constructed by combining the NK fitness landscapes and the crowdsourcing framework of “Task-Crowd-Process-Evaluation”.