Water Journal : Water Journal May 2011
water MAY 2011 85 refereed paper demand management behaviour change in the community and thus limited opportunity to further reduce consumption. Traditional economic approaches indicated a greater reduction in demand, but were unable to model the critical behavioural components that SimulAIt showed significantly affected the outcome, and thus fell short in accurately predicting consumption in this human- centric system. Conclusion ISD's agent-based SimulAIt consumer modelling and forecasting platform was used to create a highly detailed micro- simulation of over 40,000 households in Ballarat. The model simulated individual people within households, how they made decisions in the house and garden, and how their decisions were influenced by different policies and communication signals, such as water restrictions, price increases, marketing campaigns and the media. The model integrated a wide range of qualitative and quantitative social, economic, environmental and political data. Extensive validation showed greater than 95% accuracy and tracked retrospectively the complex trends over eight years. Validation also showed that the model was transferable and could forecast bounce-back when applied to the Bendigo region. Future forecasts indicate that the "bounce-back" in water consumption from easing water restrictions is minimal, peaking below pre-restriction levels, due to water-saving behaviours of consumers being maintained as a result of the long period in which restrictions have been imposed. Forecasts also showed that demand is relatively insensitive to price increases. This is due to a high level of persistent behaviour change by consumers, and thus limited opportunity to further reduce consumption. Traditional economic analysis is unlikely to model these critical behavioural components, and thus fell short in accurately predicting consumption in this human-centric system. CHW received a range of benefits from the forecasting model, including: • Greater accuracy in sales and financial forecasts to inform short- and long- term plans; • More rigorous business case to industry pricing regulators; • Ability to assess and inform future water policies, programs and marketing; • Ability to understand and quantify the benefit of specific influences or strategies in a campaign; • Identification of demand characteristics of different consumers (regions and demographics), to enable better targeting of future programs (micro-marketing); • Greater understanding and quantification of the effectiveness and benefits of past programs and strategies; • Can better inform future conservation measures and policies. Acknowledgements The authors would like to acknowledge the contribution of a wide range of people and agencies including the Victorian Department of Sustainability and Environment (DSE) for funding and support for the project. ISD acknowledges Central Highlands Water staff for assisting with data collection and providing feedback, as well as staff at Coliban Water, Yarra Valley Water, South East Water, City West Water and Barwon Water for data and knowledge provided. ISD also thanks Peter Guttmann, Helen Delaporte and Les Walker from DSE for providing useful insights and critical feedback. The Authors Dr Don Perugini (email: donperugini@ intelligentsoftware.com.au) has a PhD in Software Engineering and Computer Science from the University of Melbourne and spent 11 years at the Defence Science and Technology Organisation. He is the Managing Director and one of the founders (along with Dr Michelle Perugini) of Intelligent Software Development (ISD), Adelaide SA. Brendon Clarke is the Coordinator of Demand Management and John Frdelja is a hydrogeologist at Central Highlands Water (CHW), Ballarat VIC. References ABS Census Data, 2006: Australian Bureau of Statistics. Athanasiadis IN, Vartalas P & Mitkas PA, 2004: DAWN: A platform for evaluating water-pricing policies using a software agent society. International Environmental Modelling and Software Society 2004 International Congress. Complexity and Integrated Resources Management, p. 42. iEMSs. Berger T, 2001: Agent-based spatial models applied to agriculture: a simulation tool for technology diffusion, resource use changes and policy analysis. Agricultural Economics 25 (2001), pp. 245--260. Bonabeau E, 2002: Predicting the unpredictable. Harvard Business Review, March 2002. Dietz T, Gardner GT, Gilligan J, Stern PC & Vandenbergh MP, 2009: Proceedings of the National Academy of Sciences (PNAS), November 3, 2009. Farmer JD & Foley D, 2009: The economy needs agent-based modelling. Nature, Vol. 460, 6 August 2009. Janssen MA & Jager W, 2001: Fashion, habits and changing preferences: Simulation of psychological factors affecting market dynamics. Journal of Economic Psychology 22 (2001), pp. 745--772. Markose S, Arifovic J & Sunder S, 2007: Advances in experimental and agent-based modelling: Asset markets, economics networks, computational mechanism design and evolutionary game dynamics. Editorial, Journal of Economic Dynamics & Control 31 (2007), pp. 1801--1807. North MJ, Macal CM, Aubin J St. et al., 2010: Multi-scale agent-based consumer market modelling. Complexity Vol. 15, Issue 5, May/ June 2010, pp. 37--47. Nuttal WJ, Zhang T, Hamilton DJ & Roques FA, 2009: Second International Symposium on Engineering Systems, MIT, Cambridge, Massachusetts, June 15--17, 2009. Perugini D, Perugini M & Young M, 2008: Water Saving Incentives: an Agent-Based Simulation Approach to Urban Water Trading. SimTecT 08, Melbourne. Rixon A, Moglia M & Burn S, 2007: Exploring water conservation behaviour through participatory agent based modelling. Topics on Systems Analysis for Integrated Water Resource Management. Elsevier. Schenk TA, Loffler G, Rauh, Jurgen, 2007: Agent-based simulation of consumer behaviour in grocery shopping on a regional level. Journal of Business Research 60 (2007), pp. 894--903. Twomey P & Cadman R, 2002: Agent-based modelling of customer behaviour in the telecoms and media markets. Info Journal, 4, 1, 2002, pp. 56--63. Willis R, Stewart RA, Talebpour MR, Mousavinejad A, Jones S & Giurco D, 2009: Revealing the impact of socio-demographic factors and efficient devices on end use water consumption: Case of Gold Coast, Australia. 5th IWA Specialist Conference on Efficient Use and Management of Urban Water. Zhang T & Zhang D, 2007: Agent-based simulation of consumer purchase decision- making and the decoy effect. Journal of Business Research 60 (2007), pp. 912--922.
Water Journal April 2011
Water Journal July 2011