Received: 12 May, 2007. Accepted: 3 October, 2007.

SUMMARY

In poor regions, lack of local monetary circulation is one of the key elements causing underdevelopment. The more incoming money is passed from hand to hand, the more the local economy will be stimulated. However, in most poor areas money is spent outside the community before circulating locally, reducing the effectiveness of money inflow dramatically.
Development programs would increase their effectiveness if knowledge was available on how spending money could lead to optimized and prolonged local circulation. To gain this knowledge a simulation tool will be created, which is able to analyze financial flows, to evaluate the potency of specific actions aimed on local development, and to monitor a development scheme during the execution phase.
The basic model will be developed through a multi-agent approach, where each agent represents one (or more) family/households belonging to one of several socio-economic groups. A Social Accounting Matrix (SAM) of the local economy will be used as a basis to set up a spendings matrix for each agent, defining its spending priorities. Artificial Intelligence techniques will be used to give the agent the possibility to make decisions on how to satisfy these spending priorities. Also, social dynamics, the simulation of strategic planning behavior, learning, and exchange in limited networks will be addressed.
The simulation application will consist of a common user interface allowing the user to "play" the simulation. This user interface layer will be "pluggable" with the underlying programming layer responsible for the calculations on the simulation, so that different plug-ins may be used for different simulation techniques.

KEY WORDS

multiplier effect, simulation, multi-agent based simulation, social accounting matrix, artificial intelligence techniques

CLASSIFICATION

ACM:	J4
JEL:	E51, O18, R15