The second part of this book provides meaningful examples of agent-based modelling applications. Ranging from warfare strategies to natural resources management, the different chapters display two characteristics that, too often, are lacking in descriptions of computer simulation. First, the context of application is carefully presented, as it seems essential to understand the reasons why decision-makers decided to fund or to participate in such projects. Then, the corollary is to give the opportunity to decision-makers or practitioners to comment on the usefulness of the approach. Hence, each chapter is introduced by a specific foreword from a personality who was actively involved or interested in using the outcomes of the modelling process.
Effective applied research doesn’t need only to be deeply rooted in a context and to be accessible to stakeholders, it also needs to participate in the global backing of national capacities. In Australia, the National Research Priorities (NRPs) provide a general framework guiding research institutions. Four NRPs are currently developed: an environmentally sustainable Australia; promoting and maintaining good health; frontier technologies for transforming Australian industries; and safeguarding Australia.
One could argue that agent-based modelling and computer simulation are breakthrough sciences that deserve to fall entirely into the third priority. On the contrary, the variety of contexts discussed in this book demonstrate the large range of applications of these methods by linking each chapter of the second part of this book to a relevant NRP.
In chapter 12, Anne Dray and colleagues describe how they used multi-agent simulations in conjunction with a role playing game to develop a Negotiation Support System for groundwater management in Tarawa (Republic of Kiribati). Their Companion Modelling approach relies on three successive stages. First, a Global Targeted Appraisal focuses on social group leaders in order to collect different standpoints and their articulated mental models. Then, these contrasted models are merged into a single conceptual one using UML (Unified Modelling Language) formalism. This conceptual model is further simplified in order to create a role-playing game. This computer-assisted game is played during iterative sessions, generating innovative rules and water management scenarios among stakeholders.
Ryan McAllister and colleagues (chapter 14) present an interesting reflection on Australian pastoral land-use systems that have been characterised by private-property regimes, which, to varying degrees, have created fragmented and disconnected landscapes. The authors argue that there are both environmental and economic risks associated with productive land fragments being too small, and these risks necessitate an understanding of fragmentation’s driving forces. Understanding these forces, however, is made difficult because the problem involves social, economic and environmental factors, interacting over a range of temporal and spatial scales. Hence, the authors developed an agent-based model to explore rangeland dynamics that involve such complexity. The current model contains pastoralists, livestock, key ecological processes, and governance.
In chapter 13, David McDonald and colleagues propose a Management Strategy Evaluation (MSE) framework to demonstrate practical science-based methods that support integrated regional planning and management of coastal and marine ecosystems of the North West Shelf region (Western Australia). Their multiple-use MSE has, so far, focused on four sectors: oil and gas, conservation, fisheries and coastal development. For each sector a selection of development scenarios, provided by the relevant interest groups, is represented. These scenarios include prospective future sectoral activities and their impacts, and the sectoral response to management policy and strategies. According to the authors, the agent-based modelling software InVitro is well placed for analysing prospective social and ecological impacts of multiple-use management strategies in a risk-assessment framework such as MSE.
Katherine Daniell and colleagues (chapter 7) present a model that was used to evaluate the plausible futures of Christie Walk housing development in inner-city Adelaide, Australia. The authors underline the fact that assessing the sustainability of development proposals has become of great importance to policy and decision makers. However, effective methods of assessing the overall sustainability of housing developments (proposed or existing) have yet to be established. This chapter presents a new methodology to assess the sustainability of housing development systems. The methodology uses a Sustainability Scale approach for livelihood indicators, coupled with multi-agent simulations to represent the complex housing patterns.
In chapter 11, David Batten and George Grozev describe the development of an agent-based simulation model that represents Australia’s National Electricity Market (NEM) as an evolving system of complex interactions between human behaviour in markets, technical infrastructures and the natural environment. This simulator, named NEMSIM, is the first of its kind in Australia. Users will be able to explore various evolutionary pathways of the NEM under different assumptions about trading and investment opportunities, institutional changes and technological futures, including alternative learning patterns as participants grow and change. Simulated outcomes can help the user to identify futures that are eco-efficient—for example, maximising profits in a carbon-constrained future.
Lisa Elliston and Steve Beare (chapter 9) present a model investigating the impact of a potential incursion of Karnal bunt in wheat in a valuable agricultural producing region of Australia. An incursion management model was developed to estimate the regional economy effects of the exotic pest incursion in the agricultural sector. By developing an agent-based spatial model that integrates the biophysical aspects of the disease incursion with the agricultural production system and the wider regional economy, the model can be used to analyse the effectiveness and economic implications of alternative management strategies for a range of different incursion scenarios.
In chapter 10, Pascal Perez and colleagues have explored the complexity of illicit drug markets in Melbourne. The intricacy of multiple interactions between individuals, the various time lines linked to different aspects of harm reduction, and contrasted social rationalities observed among field practitioners (prevention, law enforcement, harm reduction) contribute to the creation of complex and unpredictable systems. In order to explore this complexity, an agent-based model called SimDrug was designed. The prototype model includes users, dealers, wholesalers, outreach workers and police forces. In order to overcome the limited knowledge we have of underground illicit activities, a trans-disciplinary group of experts regularly informed (and validated) the model with consensual rules.
Ang Yang and colleagues (chapter 8) bring us into the complex world of warfare for which a number of Agent-Based Distillation Systems (ABDs) have been developed and adopted to study the dynamics of warfare and gain insight into military operations. According to the authors, these systems have facilitated the analysis and understanding of combat. However, these systems are unable to meet the new needs of defense, arising from current practice of warfare and the emergence of the theory of Network Centric Warfare (NCW). The authors propose a network centric model which provides a new approach to understand and analyse the dynamics of both platform centric and network centric warfare.