May 18-21, 2014

Denver, Colorado, USA

SIGSIM PADS 2015 Abstracts

The abstracts of the presentation-only papers are given here.

High-Fidelity Battlefield Simulation Support Military Simulation Simulations Prepare Marine Corps for War Virtual Reality Simulation Virtual Reality Driving Simulator Virtual Reality Medical Simulator
themed object
ACM SIGSIM Conference on Principles of Advanced Discrete Simulation (PADS)
get in touch

SIGSIM PADS 2015 Abstracts

Bridging Relevance with Rigour in the Policy Modelling of Political Participation

Bruce Edmonds (Centre for Policy Modelling, Manchester Metropolitan University, UK), Luis Fernandez Lafuerza (Department of Theoretical Physics, University of Manchester, UK), Louise Dyson (Department of Theoretical Physics, University of Manchester, UK), Laurence Lessard-Phillips (Institute for Social Change, University of Manchester, UK), Ed Fieldhouse (Institute for Social Change, University of Manchester, UK), and Alan McKane (Department of Theoretical Physics, University of Manchester, UK)


In order to achieve meaningful policy analysis using simulation technology, one has to get social and formal/computer science to work well together. This is often a problem — social and formal scientists speak different languages and have radically different methodologies/concerns. However they also have necessary and complementary areas of expertise: relevance and rigour respectively. Often, when collaboration is attempted the social scientists are appalled at simulation models developed for them, being aware of the disconnect between the complexity of what they know to happen at the micro-level with the abstraction of the models. Similarly formal/computer scientists are often appalled at the lack of rigour in the models that social scientists use, often as a result of the complexity and complication of their view of the phenomena.

This presentation describes a methodology that attempts to square this circle. It uses a progressive but gradual model simplification. Here a complex simulation, which aims to integrate available social science evidence into a single consistent but dynamic ‘picture’, is constructed – a Data Integration Model (DIM). This model is typically highly complex, but well grounded in evidence from the micro-level. This DIM is then, itself, used as a target of modelling – that is a (simpler) model of the DIM is produced, but one that is easier to analyse and understand. This ‘staging’ of the abstraction process via a DIM is an implementation of the ‘KIDS’ modelling approach (Keep it Descriptive Stupid) – it contrasts with that of ‘brave’ abstraction to a simpler model, attempted in a single leap. The staged approach has several advantages: (a) one does not have to make so many a priori decisions as to what to include in the simple model (since on can eliminate some via experimentation on the DIM) (b) the simpler model is a check on the implementation of the DIM, and (c) the DIM can be used to test the assumptions behind the simpler model and its scope. The main disadvantage is the larger effort required to develop two models and check their correspondence.


Agent-Based Modelling for Exploring Policies in Complex Socio-Environmental Systems

Gary Polhill, Jiaqi Ge, Alessandro Gimona, and Nick Gotts (James Hutton Institute, UK)


Agent-based modelling has a long history of policy exploration, with perhaps one of the most popularly cited examples being Lansing and Kremer’s (1993) work on the role of Balinese water temples in managing rice paddy irrigation. Pest dynamics motivate farmers to plant at the same time, whilst water scarcity means there is a motivation to plant at different times. Lansing and Kremer were able to show that the spatial distribution of water temples in the river basin corresponded well with the spatial scale of co-ordination needed to optimally balance these conflicting influences on planting schedules. This evidence was influential in persuading policymakers that there was no need to encourage farmers to use artificial pesticides and fertilisers.

Our work at The James Hutton Institute has used agent-based modelling to explore various policy scenarios. By coupling an agent-based model of farm business decision-making with a spatially explicit model of species occupancy dynamics, we have been able to demonstrate that outcome- rather than activity-based agri-environmental incentive schemes might be more robustly capable of maintaining biodiverse landscapes. (Polhill et al. 2013) The model uses a decision-making algorithm from the Artificial Intelligence literature, based on psychological evidence about how experts make decisions (Aamodt and Plaza 1994). Our results, based on several thousand runs, show that below a threshold in government expenditure, market dynamics are the main driver of biodiversity outcome, and only under the most benign circumstances with respect to price stability, input costs and farm business agent aspirations are high landscape-scale species richness observed. Above that level, outcome-based incentive schemes robustly deliver ‘good enough’ biodiversity, whilst the results of activity-based schemes have further sensitivities to market dynamics and government expenditure.

In work funded by the European Commission Seventh Framework Programme, we have simulated influences on domestic energy consumption using an empirically calibrated agent-based model (Gotts et al. 2014) based on psychological theory about pro-environmental decision-making (Lindenberg and Steg 2007). In this model, we are interested in how policy influences domestic energy expenditure in the context of other influences: appliances owned by friends and neighbours, and fuel price and salary inflation. The policies considered were removing inefficient appliances from the marketplace, and providing financial incentives for insulation and installation of condensing boilers. Preliminary results from the model have suggested that although these policies have an effect, it is dwarfed by that of fuel price and salary inflation (Gotts and Polhill 2014).

More recently, also funded by the European Commission, we have been considering the relationship between work-life balance and CO2 emissions in the context of a model of commuting in Aberdeen. Contrasting the influence of introducing flexible working hours in employees’ contracts and building a bypass to ease congestion during peak demand, we find that greater savings in mean commuting time can be achieved with even a small amount of flexibility than with a bypass. In the case of CO2 emissions, our early results suggest that the addition of a bypass will slightly increase emissions, whilst flexible working hours achieves a small reduction, with greater flexibility halving the peak emissions during rush hour (Ge and Polhill subm.)


This work has been funded by the Scottish Government Rural Affairs and the Environment Portfolio Strategic Research Theme 1 (Ecosystem Services), and by the European Commission Seventh Framework Programme grant agreement nos. 225383 (GILDED) and 613420 (GLAMURS).


  • Aamodt, A. and Plaza, E. (1994) Case-based reasoning: foundational issues, methodological variations, and system approaches. AI Communications 7 (1), 39-59.
  • Ge, J. and Polhill, G. (subm.) An agent-based model of daily commute in Aberdeen, UK. Submission to the Social Simulation Conference SSC 2015, Groningen, 14-18 September 2015.
  • Gotts, N. M., Polhill, G., Craig, T. and Galan-Diaz, C. (2014) Combining diverse data sources for CEDSS, an agent-based model of domestic energy demand. Structure and Dynamics: eJournal of Anthropological and Related Sciences 7 (1),
  • Gotts, N. and Polhill, G. (2014) The CEDSS model of direct domestic energy demand. Social Simulation Conference SSC 2014, Barcelona, 1-5 September 2014.
  • Lansing, J. S. and Kremer, J. N. (1993) Emergent properties of Balinese water temple networks: coadaptation on a rugged fitness landscape. American Anthropologist 95 (1), 97-114.
  • Lindenberg, S. and Steg. L. (2007) Normative, egoistic and hedonic goal-frames guiding environmental behaviour. Journal of Social Issues 63, 117-137.
  • Polhill, J. G., Gimona, A. and Gotts, N. M. (2013) Nonlinearities in biodiversity incentive schemes: A study using an integrated agent-based and metacommunity model. Environmental Modelling and Software 45, 74-91.


Multi-Agent Simulation and Animation with PreSage-2

Jeremy Pitt (Imperial College, UK)


PreSage-2 is an open source multi-agent simulation and animation platform, designed for simulation of large-scale agent-based systems, with agents of arbitrary complexity whose behaviour is regulated by an external, conventional, and mutable set of rules. This talk will give an overview of the architecture, programming method and experimental features of PreSage-2, and then highlight some of its applications, including policy modelling of the Kyoto protocol, self-organised resource allocation, and a serious game for Smart Grids.



slide up button