The significant objective of this paper is to describe how Cellular Automata (CA) can be used to predict land use land change in the settlement growth and show the advantage of integrating the Neural Network with CA. Cellular automata have been utilized as a prediction technique in the study of an impressively wide range of dynamic phenomena. A spatial simulation model comprises an assortment of processes performed on spatial data that will produce information, by and large in the form of a map. CA models exude superior performance in simulating land changes compared to conventional models .
CA are much simpler than complex mathematical equations and produce results that are more meaningful and useful. Temporal and spatial complexities of systems can be efficiently modelled by precise definition of transition rules in CA models. GIS is a technology that is employed to view and analyse data from a geographic perspective. In practice, grid cells covers the selected area of study. Consequently, specific ground surface attribute value of interest occurring at the centre of each cell point is recorded as the value for that cell.

Simulation is carried out using traditional CA and Neural Network based CA for the settlement growth for an artificial city and real city. A number of land use sprawl parameters, and different size and shape of neighborhood with some testing constraints are used in this simulation. The results reveal that neural network CA method is more appropriate than traditional CA for predicting the settlement growth.