written by: M. Fajri
Simulation is one method to solve a problem. By doing a simulation, it can be seen the factors that influence a system, and can see changes in the system when one of the elements in the system changes. One simulation method that can be used is agent-based modeling.
Agent Based Modeling is one of the modeling methods that can simulate the interaction between agents / individuals in a system, so that it can be seen the model of the system and the effects of the individual on the system. Each of these individuals has a unique or different characteristic and behavior from one another. So that when there is an interaction between agents, there will be a behavior that cannot be predicted beforehand. The ability of agents to determine this behavior is called self-organization. Previous unpredictable behavior is called emergence behavior.
One example of a system that can be modeled with agent-based modeling is ant colonies. Ant colonies are included in a complex system, because they consist of several entities that interact with each other, each ant entity has its own characteristics or characteristics, due to the interaction of the interactions between these entities, it looks like properties or characteristics that are unexpected or far different from the characteristics of each entity. Ants have interesting characteristics when searching for food, where ants will colonize and follow the same track to find food. Many hypotheses arise about how ants look for food in the colony, such as the command of the queen ant to ants or the communication between ants (as do bees). But these two hypotheses are less precise, because what happens to ants is not communicating or being ordered.
The self-organization of the ant is that the ant will move randomly, and find food, then return to its nest. In each movement, the ant will leave the pheromone. Then will again look for food, by following the strongest pheromone. So that the ant will gradually follow the shortest path to find food, because the shortest path always has the highest pheromone From the self-organization, emergence behavior is formed, where ants will follow the strongest pheromone to find food. And other ants will also follow the shortest track, so that behavior will be formed as if the ants were communicating to find food, thus finding the shortest route.
Parunak, Wilensky, and colleagues (Parunak et al., 1998; Wilensky, 1999b; Wilensky & Reisman, 2006) explain that there are several differences between Agent Based Modeling and other modeling. First, Agent Based Modeling can describe heterogeneous populations, or consider agents having different properties, while in most other models the population is considered homogeneous. Second, Agent Based Modeling does not require knowledge of phenomena as a whole, which does not require knowing the pattern as a whole but only knowledge of individual behavior. Furthermore, Agent Based Modeling is closer to the real system, because it models the behavior of each different individual. Finally, Agent Based Modeling modeled in more detail, using the “bottom-up” method, so that it can see individual levels and aggregate levels at the same time.
Software yang digunakan untuk Agent Based Modelling ada beragam. Beberapa software yang dapat digunakan adalah Repast, Starlogo, Netlogo, C, C++, Java atau Pascal. Dari beberapa software tersebut, software yang sering digunakan adalah Netlogo. Hal ini dikarenakan Netlogo menggunakan bahasa pemprograman yang relative mudah dan mampu memperlihatkan tampilan yang mudah dipahami.
There are many types of software used for Agent Based Modeling. Some software that can be used is Repast, Starlogo, Netlogo, C, C ++, Java or Pascal. From these software, the software that is often used is Netlogo. This is because Netlogo uses relatively easy programming language and is able to show an easy-to-understand appearance.
Netlogo is a software that can be used to model a system and can give orders to its agents so that it can describe the nature of different agents and form a complex system. NetLogo was first developed in 1999 by Uri Wilensky and has since been developed at the Center of Connected Learning and Computer-Based Modeling. Users can give special orders to thousands of agents at once and can model interactions that can occur so that they can describe the conditions of a complex system.
NetLogo runs using a Java virtual machine, so it can be operated on all operating system platforms such as: Mac, Windows, Linux, and so on. In NetLogo there are four types of agents, namely:
- Turtles, are agents that move in the environment and can interact with patches or other agents.
- Patches, where the agent moves. A collection of these agents will form the environment or environment.
- Link, is a link between two turtles in the environment.
- Observer, is an external agent that can manipulate all agents in the environment. The characteristic for knowing a procedure in a program made from an observer’s point of view is the ASK command at the beginning of a command. Observer cannot be seen visually in the environment and will carry out orders according to the user’s decision.