NEURAL-NETWORK MODEL FOR CONSEQUENCE ANALYSIS OF DEVELOPMENTAL PROPOSALS

Application of neural networks for regional system identification is r eported in the context of sustainable development. Neural network I ca ptures relationships among natural resource status, economic condition s, amenities, and environmental status; it is used for the prediction of environmental status resulting from the socioeconomic developmental alternatives. Neural network II predicts the quality of life levels. A geographic information system linked with an external relational dat abase system maintained spatial and numerical data, and provided input s to neural networks. The integrated system is used for consequence an alysis of alternate developmental scenarios, and for the choice of a p referred scenario. The case of developmental planning in the national capital region of India is presented to illustrate the effectiveness o f this modeling approach.