lfbraz/mas-mbti-model

This repo is used to develop a multiagent-system model to simulate the interactions between the agents with different personality types.

A Buyer-Seller approach using an MBTI agent-based model

This repository has the code used to create agent-based models developed with personality types inspired by MBTI (Myers-Briggs Type Indicator).

For more details on the theory behind the implementations and the Buyer-Seller approach, please see some of the published papers:

• Using MBTI Agents to Simulate Human Behavior in a Work Context - Springer
• Using the Myers-Briggs Type Indicator (MBTI) for Modeling Multiagent Systems - Revista de Informática Teórica e Aplicada
• Extending BEN architecture for modeling MBTI agents - HAL Open Science
• Simulating Work Teams using MBTI agents - MABS 2022 – The 23rd International Workshop on Multi-Agent-Based Simulation
• Um estudo do Myers-Briggs Type Indicator (MBTI) para modelagem de sistemas multiagentes no apoio a processos de recrutamento e seleção nas empresas (Best Papers) - 14th WESAAC - UTFPR

Prerequisites

• GAMA Platform - V1.8.2

How to use?

• Clone this repo and open the mbti-buyer-seller folder.
• You can check the buyer-seller model to explore how to use the MBTI inspired model, or you can simply import the mbti.gaml file into your own model.

How to run using Headless Mode?

• Run the following command in your headless folder: gama-headless.bat "<PATH-OF-YOUR-LOCALWORKSPACE>\headless_input\gamaDefaultMBTI.xml" /output

The gamaDefaultMBTI.xml file has an example of how to run 10 simulations (nine with the same SEED and another one with a different seed) using headless mode passing some parameters (seed value is inherited from the XML file). You can find gama-headless.bat in the Gama Program's File folder (usually C:\Program Files\Gama\headless if you use Windows).