Simulation modelling and analysis is the process of creating and experimenting with a computerised mathematical model (Chung, 2004) imitating the behaviour of a real-world process or system over time (Banks, 1998). Simulation is used to describe and analyse the behaviour of a system when asking ‘what-if’ questions about the real system and aid in the design of real systems. The main objectives are: (Pedgen et. al., 1995)
• Gaining insight into the operation of a system
• Developing operating or resource policies to improve system performance
• Testing new concepts and/or systems before implementation
• Obtaining information without disturbing the actual system
Simulation modelling is a versatile technique well suited for the study of some complex problems, to tackle previously untouched, often apparently unmanageable problems. Simulation is often the obvious tool to be tried.
Simulation modelling is specifically useful for policy makers and strategic management, gaining insight into general future developments. The people involved are decision makers, researchers with knowledge of the situation being modelled, and software developers.
Approach
Different types of modelling and simulation techniques exist and it is not possible to detail them here. What follows is a list of general steps upon which all models rely.
1. Scope and set up of model: First the study is defined, objectives are inventoried and the model is set up, including general assumptions on relevant factors, either variables or constants and how they are related.
2. Data collection: Simulation modelling stands or falls on the availability, applicability, and reliability of the data: garbage in, garbage out (GIGO) applies.
3. Model testing: The relevant data is entered into to the model and calculated. Outcomes are compared to reality, such that the model is validated. Possibly, some factors are calibrated such that outcomes are more realistic.
4. Analysis: Finally, the model can be used to change some factors, either ‘what if’-scenarios or predicted changes for example based on extrapolation.
Resources & outputs
Resources depend, among other things, on the system definition, model and data availability, level of detail, level of uncertainty, system complexity and research questions. Simulation modelling requires specific knowledge skills for model building and analysis. Typical outputs are reports with explanations and interpretations of the practical situation, the model, and its outputs.
Pros and cons
The main benefits of simulation modelling are (Chung, 2004):
• Experimentation in limited time
• Reduced analytical requirements
• Easily demonstrated models
The main limitations are (Chung, 2004):
• Simulation cannot give accurate results when the input data are inaccurate
• Simulation cannot provide easy answers to complex answers
• Simulation cannot solve problems by itself
“The variables and data chosen for the model are still subjective, although the calculations suggest objectivity.” and “It is often non-transparent and difficult to explain, what the model does and how it is calculated.” (Kerstin Cuhls, 2005)
Agent Modelling
ABM is a computational method to create, analyse and simulate models including individual agents interacting within an environment. Agents in ABM can be simply defined as autonomous (individual and collective) decision-making entities. ABMs combine elements of game theory, complex systems, emergence, computational sociology, multi-agent systems, and evolutionary programming. Monte Carlo Methods are used to introduce randomness. The models simulate the simultaneous operations and interactions of multiple agents, in an attempt to re-create and predict the appearance of complex phenomena.
Individual agents are typically characterized as bounded rational, presumed to be acting in what they perceive as their own interests, such as reproduction, economic benefit, or social status, using heuristics or simple decision-making rules. ABM agents may experience ‘learning’, adaptation, and reproduction.
The advantage of ABM over other modelling approaches is that it is not constrained by equilibrium assumptions; it can include agents with diverging preferences and market behaviours, and account for a greater degree of the unpredictability that springs from complex social interaction.
Their application in the field of future studies is so far limited, yet interested in the field is increasing dramatically, given the flexibility of the tool. Indeed, agent-based models have been used since the mid-1990s to solve a variety of business and technology problems. Examples of applications include supply chain optimization and logistics, modelling of consumer behaviour, including social network effects, distributed computing, and workforce management.
Agent-based modelling tools can be used to test how changes in individual behaviours will affect the system’s emerging overall behaviour. Other models have analysed the spread of epidemics, biological applications including population dynamics, the growth and decline of ancient civilizations, evolution of ethnocentric behaviour, forced displacement/migration, language choice dynamics.