Developping 2D and 3D multiagent geosimulation, a method and its application: the case of shopping behavior geosimulation in Square One Mall (Toronto)

Liste des illustrations

• Fig. 2.‎1. The methodological process of (Drogoul et al., 2002).
• Fig. 2.2. Generic stages in the simulation research process of (Ramanath and Gilbert, 2003).
• Fig. 3.1.The main steps of our proposed method.
• Fig. 3.‎2.The simulation Passive Agent (PA).
• Fig. 3. 3.The simulation Active Agent (AA).
• Fig. 4.1.Step 1: identification of the application domain, the geosimulation end-users, and their needs.
• Fig. 4.2. Step 2: identification of the characteristics of the phenomenon to be simulated and its environment.
• Fig. 4.3. The hierarchy of needs proposed by (Maslow, 1954).
• Fig. 4.4. The processes of human consumption behavior (Duhaine et al., 1996)
• Fig. 4.5. The consumer decision-making model proposed by (Engel eet al., 1968).
• Fig. 4.6. The consumer decision-making model proposed by (Howard ans Sheth, 1969).
• Fig. 4.7. The consumer decision-making model proposed by (Nicosa, 1966).
• Fig. 4.8. Our generic model of the shopping behavior process.
• Fig. 4.9. The environment, context, and situation (Belk, 1974).
• Fig. 4.10. The model of the decision-making process under the pressure of time presented by (Dhar and Nowlis, 1999).
• Fig. 4.11. The characteristics of shopping behavior in a mall.
• Fig. 5.1. Step 3 of our proposed method.
• Fig. 5.2. The 2D spatial structure of the agents belonging to the physical environment (the mall).
• Fig. 5.3. The 3D spatial structure of the agents belonging to the physical environment (the mall).
• Fig. 6.1. Collect data using a survey.
• Fig. 6.2. Collect and analyze the data used as input for the multiagent geosimulation.
• Fig. 6.3. The GIS (2D), corresponding to the first floor of Square One.
• Fig. 6.4. The 3D model of the first floor of Square One (in a 3D viewer).
• Fig. 6.5. The structure of the individual questionnaire.
• Fig. 6.6. The structure of the individual questionnaire (Specific for Square One mall).
• Fig. 6.7. The structure of the group questionnaire.
• Fig. 6.8. A screenshot to input non-geographic data (demographic data).
• Fig. 6.9. A screenshot to input geographic data (the shoppers’ itineraries).
• Fig. 6.10. Gender of the questionnaire respondents.
• Fig. 6.11. Age group of the questionnaire respondents.
• Fig. 6.12. The general objectives of shoppers (planned).
• Fig. 6.13. The shopping period in the day.
• Fig. 6.14. The parking variable (first floor of Square One mall).
• Fig. 6.15. The parking variable (second floor of Square One mall).
• Fig. 6.16. The parking variable (first floor of Square One mall) (cartographically).
• Fig. 6.17. The parking variable (second floor of Square One mall) (cartographically).
• Fig. 6.18. The entrance door variable (first floor of Square One mall).
• Fig. 6.19. The entrance door variable (first floor of Square One mall) (cartographically).
• Fig. 6.20. Combination of gender and age group variables.
• Fig. 6.21. Combination of gender and entrance door variables (door 0).
• Fig. 6.22. Combination of gender, age group, and entrance door variables (door 0).
• Fig. 7.1. Select the multiagent geosimulation tool/platform/language.
• Fig. 7.2. Create and execute the multiagent geosimulation models: Geosimulation prototype.
• Fig. 7.3. The two modules of the MAGS platform.
• Fig. 7.4. The MAGS interface of simulation specification.
• Fig. 7.5. States of the agents in the MAGS platform.
• Fig. 7.6. The tree of objectives that composes the behavior of an active agent in MAGS.
• Fig. 7.7. The structure of an objective in MAGS’s formalism.
• Fig. 7.8. A screenshot of the GIS of Square One mall’s first floor.
• Fig. 7.9. The TextureMap.
• Fig. 7.10. The ObstacleMap.
• Fig. 7.11. The AriadneMap.
• Fig. 7.12. The AgentsMap.
• Fig. 7.13. The 3D model of Square One mall’s first floor.
• Fig. 7.14. The screen of specification, of geographic data in the simulation environment.
• Fig.7.15. The profile definition of an agent in the MAGS platform.
• Fig. 7.16. The specification of an agent’s static states in the MAGS platform.
• Fig. 7.17. The specification of an agent’s dynamic states in the MAGS platform.
• Fig. 7.18. The window of the simulation scenario specification.
• Fig. 7.19. The hierarchy of shopping activity patterns in the MAGS platform.
• Fig. 7.21.The 2D simulation in the MAGS platform (Square One mall).
• Fig. ‎07.22.The 3D simulation in the MAGS platform (Square One mall).
• Fig. 7.23.The 3D simulation in the MAGS platform (Square One mall).
• Fig. 8.1. Collect and analyze the output generated by the geosimulation prototype.
• Fig. 8.2. The usage of shopping observer agents to collect data about the virtual shoppers.
• Fig. 8.3. An observer agent and some virtual shopper agents in 3D mode in the MAGS platform.
• Fig. 8.4. A 3D simulation of an observer agent that observe virtual shoppers.
• Fig. 8.5. The positions of the observer agents in the virtual mall (Square One mall).
• Fig. 8.6. The structure of files storing the geosimulation output data (n observers).
• Fig. 8.7. Coupling of the geosimulation prototype and our analysis tool.
• Fig. 8.8. The analysis files’ structures.
• Fig. 8.9. The simulation analysis results for the frequentation of Square One by 390 virtual shoppers.
• Fig. 8.10. The selection of the observer and the variables to be analyzed.
• Fig. 8‎.11. The analysis results for visitation of the corridor related to observer X.
• Fig. 8.12. The frequentation analysis results for corridor X.
• Fig. 8.13. The frequentation analysis results for corridor Y.
• Fig. 8.14. The analysis for the combination of the gender/age_group and the corridor.
• Fig. 8.15. Details of the analysis combining gender/age_group and the corridor.
• Fig. 8.16. Coupling the shopping behavior geosimulation and the OLAP/SOLAP tool.
• Fig. 8.17. The structure of the database that contains the simulation output data.
• Fig. 8.18. Distribution of software shopper agents that are between 51 and 65 years old, for five major stores (Wall-Mart, Sears, Zellers, Old Navy, and The Bay).
• Fig. 8.19. Distribution of software shopper agents for five major stores. The table shows the number of shoppers per age group for each store. The map and bar chart show the number of shoppers that are between 18 and 25.
• Fig. 8.20. Distribution of software shopper agents for five major stores. The bar chart shows the number of shoppers per age group for Sears and Wall-Mart.
• Fig. 8.21. Cartographic representation of the distribution of shopper agents, comprising two maps (one for male and another for female agents), that includes superimposed pie charts to present the distribution of each age group.
• Fig. 8.22. Cartographic representation of the distribution of software agents, comprising one map for each age group, and superimposed pie charts to represent gender.
• Fig. 8.23. Cartographic representation of the distribution of software agents, comprising one map for each combination of age group and gender.
• Fig. 8.24. Cartographic representation of the distribution of software agents by age group
• Fig. 9.1. Verify and validate the multiagent geosimulation models
• Fig. 9.2. Test and document the multiagent geosimulation
• Fig. 9.3. A real plan of the first floor of Square One mall.
• Fig. 9.4. The 2D model of the virtual mall.
• Fig. 9.5. The 3D model of the virtual mall.
• Fig. 9.‎6. The positions of agents A and B in the simulation environment (Situation 1).
• Fig. ‎9.7. Agent A visits shop 64.
• Fig. 9.8. Agent B visits shop 134.
• Fig. 9.9. The locations of agents A and B in the simulation environment (Situation 2).
• Fig. ‎9.10. Agent A does not visit shop 134 because it does not perceive it.
• Fig. 9.11. The positions of agents A and B in the simulation environment (Situation 3).
• Fig. 9.12. Agent A visits shop 63, but agent B does not (agent B does not know shop 63).
• Fig. 9.13. The locations of agents A and B, and the restroom in the simulation environment.
• Fig. 9.14. Agent A feels the need to use the restroom, but not agent B (B goes to shop 99).
• Fig. 9.15. Agent A visits shop 99 after using the restroom.
• Fig. 9.16. Agent A reaches its shopping time limit and decides to leave the mall
• Fig. 10.1. Use the multiagent geosimulation.
• Fig. C.1. The specification of the objective Go_To_A_Shop x.
• Fig. C.2. The specification of the objective Go_To_A_Restroom.
• Fig. C.3. The specification details of the objective Leave_The_Mall.