![Modelling Individual Consumer Behaviour Email: [email_address] ; [email_address] 26 th August 2009 School of Geography FACULTY OF ENVIRONMENT Kirk Harland Alison Heppenstall](https://image.slidesharecdn.com/harland-2-090901015835-phpapp02/85/Modelling-Individual-Consumer-Behaviour-1-320.jpg)
Modelling Individual Consumer Behaviour
- 1. Modelling Individual Consumer Behaviour Email: [email_address] ; [email_address] 26 th August 2009 School of Geography FACULTY OF ENVIRONMENT Kirk Harland Alison Heppenstall
- 2. Presentation Motivation Project overview Population generation Adding behaviour Future Research School of Geography FACULTY OF ENVIRONMENT
- 3. School of Geography FACULTY OF ENVIRONMENT Motivation Retail models: mature but limited aggregate populations limited shopping behaviours (internet, multi-purpose) increased data and computational power Recent work: ABM of retail market (demand side) But: consumers still aggregated transport network simplistic (Heppenstall, et. al 2005, 2006)
- 4. School of Geography FACULTY OF ENVIRONMENT ESRC First Time Grant Oct 2008 – Oct 2010 Assess different methods of generating population Create population incorporating commercial marketing survey dataset Generate agent rule-sets Replace aggregate demand-side of model with ‘individual’ agents Optional extra… incorporate realistic road network for agents to traverse
- 5. School of Geography FACULTY OF ENVIRONMENT Model Overview
- 6. School of Geography FACULTY OF ENVIRONMENT Phase 1: Population Generation Require a method of generating robust populations tailored to specific issues flexibly and quickly. Different population required for a health study to that for a retail market etc… Compared three synthesis methods to create a population and then assessed the results using a variety of statistics: Deterministic Reweighting (Smith et al . 2009) Conditional Probabilities (Birkin & Clarke 1988;1989) Combinatorial Optimisation – Simulated Annealing (Press et al. 1992) See Harland et al. for more information
- 7. School of Geography FACULTY OF ENVIRONMENT Population Generation – Data Experiment data consists of 715,402 individuals residing in the Leeds area. Extract from the Sample of Annonymised Records, Small Area Microdata file is used as the sample population 2001 Census of Population data used for constraints (univariate constraints applied)
- 8. School of Geography FACULTY OF ENVIRONMENT Constraints MLSOA: 106 LLSOA: 476 OA: 2439 Constraint DR CP SA CE % CE CE % CE CE % CE Middle Layer Super Output Area Gender 29,510 4.12 102 0.01 0 0.00 Ethnic Group 14,897 2.08 2,290 0.32 0 0.00 Age 128,999 18.03 144 0.02 0 0.00 Marital Status 95,335 13.33 478 0.07 0 0.00 NSSEC 84,731 11.84 4,378 0.61 0 0.00 Highest Qualification 229,407 32.07 2,569 0.36 0 0.00 Lower Layer Super Output Area Gender 30,297 4.23 176 0.02 0 0.00 Ethnic Group 15,631 2.18 4,010 0.56 0 0.00 Age 131,230 18.34 245 0.03 0 0.00 Marital Status 96,453 13.48 842 0.12 0 0.00 NSSEC 88,282 12.34 9,659 1.35 0 0.00 Highest Qualification 228,425 31.93 5,219 0.73 0 0.00 Output Area Gender 33,430 4.67 245 0.03 0 0.00 Ethnic Group 16,707 2.34 5,292 0.74 0 0.00 Age 135,673 18.96 418 0.06 0 0.00 Marital Status 98,696 13.80 1,828 0.26 0 0.00 NSSEC 95,117 13.30 21,939 3.07 0 0.00 Highest Qualification 227,720 31.83 11,385 1.59 0 0.00
- 9. School of Geography FACULTY OF ENVIRONMENT Population Generation – Increasing Constraints Number of constraints increased from 6 to 15, including provision of care, tenure, health and economic activity 14,846 people misclassified. 2.08% of total population.
- 10. School of Geography FACULTY OF ENVIRONMENT Population Generation – Increasing Constraints (2) Not all constraints have the same impact
- 11. School of Geography FACULTY OF ENVIRONMENT Next Stages Creation of households from Sample Annoymised Records. Incorporation of geodemographic information and commercial data whilst providing a realistic geographical location for the synthetic residence. Future research Christmas 2009: creation of an agent-based model from the synthetic population. Application of a behavioural architecture within the agents.
- 12. School of Geography FACULTY OF ENVIRONMENT Human Behaviour PECS (Schmidt, 2000; Urban 2000) P hysical Condition E motional States C ognitive Capabilities S ocial Status Malleson et al, (2009) In press.
- 13. Burglar Agents Needs “ lifestyle”, sleep, drugs Cognitive map of environment Burglary: Decide to commit a burglary Find suitable target Visit known good target, or See opportunity on route without specifically looking, or Search for a target Burgle http://crimesim.blogspot.com
- 14. School of Geography FACULTY OF ENVIRONMENT Summary Creating individual level populations is computationally intensive with processing issues (some approaches to improve the performance of the model: threading, indexing, cursors). Acquiring geodemographic data and SARs data Other interesting things that we haven’t had time to do: further increasing the number of constraints, does this produce a more realistic synthetic population? Use synthetic individual and their characteristics to create households, rather than SARs data (bottom-up).
- 15. School of Geography FACULTY OF ENVIRONMENT References Birkin, M., and Clarke, M. 1988 , SYNTHESIS-a synthetic spatial information system for urban and regional analysis: methods and examples, Environment and Planning A , 20, pp 1645-1671 Birkin, M., and Clarke, M. 1989 , The Generation of Individual and Household Incomes at the Small Area Level using Synthesis, Regional Studies , 23:6, pp 535-548 Harland, K, Birkin, M.H., Smith, D.M. and Heppenstall, A.J., Creating realistic synthetic populations at Varying Spatial Scales: A comparative critique of population synthesis techniques. In review. Heppenstall, A.J., Evans, A.J. and Birkin, M.H. (2005), A Hybrid Multi-Agent/Spatial Interaction Model System for Petrol Price Setting. Transactions in GIS 9(1): 35 - 51. Heppenstall, A.J., Evans, A.J. and Birkin, M.H., (2006) Application of Multi-Agent Systems to Modelling a Dynamic, Locally Interacting Retail Market. JASSS. vol 9(3). Malleson, N.S., Heppenstall, A.J., and See, L.M., Simulating Burglary with an Agent-Based Model. Computers, Environment and Urban Systems. In press. Press, W. H., Teukolsky, S. A., Vetterling, W. T., and Flannery, B. P. 1992 , Numerical Recipes in C: The Art of Scientific Computing. 2 nd Edition. Cambridge University Press. Cambridge, England Schmidt, B. ( 2000). The Modelling of Human Behaviour. SCS Publications, Erlangen, Germany. Smith D., Clarke G., and Harland K. (2009) Improving the synthetic data generation process in spatial microsimulation models. Environment and Planning A . 41. pp 1251-1268 Stillwell, J. C. H., Duke-Williams, O. (2007) Understanding the 2001 UK census migration and commuting data: the effect of small cell adjustment and problems of comparison with 1991, Journal of the Royal Statistical Society 170: pp 1–21. Urban, C, (2000). PECS: A reference model for the simulation of multi-agent systems. In Ramzi Suleiman, Klaus G. Troitzsch, and Nigel Gilbert, editors, Tools and Techniques for Social Science Simulation, chapter 6, pages 83–114. Physica-Verlag.
- 16. School of Geography FACULTY OF ENVIRONMENT Model Structure – Flexible Modelling Framework
- 17. School of Geography FACULTY OF ENVIRONMENT Model Structure – Flexible Modelling Framework
- 18. School of Geography FACULTY OF ENVIRONMENT Model Structure – Flexible Modelling Framework