Three principal advantages are claimed of agent-based over traditional top-down modeling techniques. The agent-based approach:

• | captures emergent phenomena |

• | provides a natural environment for the study of certain systems, and |

• | is flexible, particularly in relation to the development of geospatial models |

Emergence is a phenomenon, along with other surprising and unexpected behaviors unfamiliar to the classical sciences, such as self-organization, chaos, adaptation, etc., which are characteristic of complex systems. Neural networks (see further, Section 8.3, Artificial Neural Networks (ANN)) provide a well known example of complex structures that are capable of organized behavior as a result of the interaction of many interconnected neurons. More specifically, the study of phenomena characterized by interactions among many distinct components can be described as ‘aggregate complexity’. Emergent phenomena are characterized by stable macroscopic patterns arising from local interaction of individual entities. By definition, emergent phenomena cannot be reduced to the system’s parts; the whole is more than the sum of the parts. Thus, emergent phenomena can exhibit properties that are decoupled (i.e. logically independent) from the properties of the system’s parts. For example, a traffic jam often forms in the opposing lane direction to a traffic accident; a consequence of ‘rubber-necking’.

Studying the behavior of collections of entities focuses attention on relationships between entities. Characteristics of emergent phenomena make them difficult to understand and predict, particularly as emergent outcomes can be counter intuitive. In summary, the purpose of aggregate complexity is to arrive at understanding by reduction, and reassembly of a system of aggregate complexity, where the critical break with previous reductionist science is the attempt at reassembly. Aggregate complexity is of particular interest to spatial analysts because it implies that the local spatial configuration of interactions affects outcomes at the whole system level.

Since agent-based models describe the behavior and interactions of a system’s constituent parts from the bottom up, they are the canonical approach for modeling emergent phenomena. Bonabeau (2002) has identified a (non-exhaustive) list of situations where agent-based models can be useful for capturing emergent behavior:

• | Complex interactions: Interaction between agents is complicated, non-linear, discontinuous, or discrete (i.e. the behavior of an agent can be altered dramatically, even discontinuously, by other agents). This can be particularly useful if describing discontinuity of individual behavior is difficult, for example, using differential equations |

• | Heterogeneous populations: The ability to design a heterogeneous population of agents with an agent-based model is significant. Agents can represent any type of unit, from which intuitive collections of individual units can be formed, from the bottom up. Unlike agent-based models, aggregate differential equations tend to smooth out fluctuations. This is important because under certain conditions, fluctuations can be amplified: a system can be stable (approximately constant or exhibiting a linear trend) but susceptible to large perturbations. Heterogeneity also allows for the specification of agents with varying degrees of rationality (see above) |

• | Topological complexity: The topology of agent interactions is heterogeneous and complex. Aggregate flow equations usually assume global homogeneous mixing, but the topology of an interaction network can lead to significant deviations from predicted aggregate behavior. This is particularly poignant for social processes, because physical or social networks matter; and, when agents exhibit complex behavior, including learning and adaptation |

• | Appropriate model framework: In many cases ABM is a natural method for describing and simulating a system composed of real-world entities. The agent-based approach is more akin to reality than other modeling approaches, rendering ABM inherently suited to simulating people and objects in very realistic ways. For example, it is arguably easier to conceptualize and model how evacuees exit a building during an emergency, than to produce equations that govern the dynamics of evacuee densities. Nonetheless, because equations regarding evacuee density result from the behavior of evacuees, the agent-based approach will also enable the user to study aggregate properties. In particular, the agent-based approach can be useful when it is more natural to describe the constituent units of a system under some or all of the following conditions: |

(i) The behavior of individuals cannot clearly be defined through aggregate transition rates (e.g. panic within a fleeing crowd)

(ii) Individual behavior is complex. Although hypothetically any process can be explained by an equation, the complexity of differential equations increases exponentially as the complexity of behavior increases. Describing complex individual behavior with equations can therefore become intractable

(iii) Agent behavior is stochastic. Points of randomness can be applied strategically within agent-based models, as opposed to arbitrarily within aggregate equations

• | Flexibility: Finally, the agent-based approach to modeling is flexible, particularly in relation to geospatial modeling. Notably, spatial simulations benefit from the mobility that agent-based models offer. To reiterate, an agent-based model can be defined within any given system environment (e.g. a building, a city, a road network, a computer network, etc.). Furthermore, agents have the ability to move within their environment, in different directions and at different velocities. Agent mobility makes ABM very flexible in terms of potential variables and parameters that can be specified. Neighborhoods can also be specified using a variety of mechanisms. The implementation of agent interactions can easily be governed by space, networks, or a combination of structures. This would be far more complex to model using mathematics, for example. Significantly, agent-based models can regulate behaviors based on interactions at a specific distance and direction. Agent-based models also provide a robust and flexible framework for tuning the complexity of agents (i.e. their behavior, degree of rationality, ability to learn and evolve, and rules of interaction). Another dimension of flexibility is the ability to adjust levels of description and aggregation. It is easy to experiment with aggregate agents, sub groups of agents, and single agents, with different levels of description coexisting within a model. Thus, the agent-based approach can be used when the appropriate level of description or complexity is unknown, and finding a suitable level requires exploration of scenarios |