摘要 |
An agent-based demand prediction system generates weather-dependent product demand predictions using agent-based modelling in which the clients for the product are the agents. The system comprises and analytics module 100, a weather module 400 and an agent-based simulation module 200. The analytics module 100 is arranged to select and/or generate agent behaviour rules based on historical data relating to weather-dependent product acquisition or purchases by client agents. The weather module 300 is arranged to acquire weather forecasts in real time and to access pre-defined weather event rules. The agent-based simulation module 200 is arranged to use the agent behaviour rules and weather forecasts to simulate agent behaviour and generate product demand predictions. The weather module 300 is arranged to compare the acquired weather forecasts with the pre-defined weather event rules, and trigger a simulation in the agent-based simulation module 200 based on a match between an acquired weather forecast and a pre defined weather event rule. The rule library may include product acquisition data that varies geographically. Competitor agent behaviour rules may also be taken into account. |