Simulating spatial growth patterns of urban areas in Ghana: An agent based modeling approach

Abstract

In Sub-Saharan Africa, rapid urban growth has outstripped the capacities of central and local government to control and regulate urban land development; hence people construct their homes on available lands most of which are liable to floods. This research set out to identify the driving forces of spatial growth and to simulate spatial growth patterns of structures in the Shama district of the Western Region of Ghana using an agent-based modeling approach. The study was conducted within a framework of Net Logo with geographic information systems (GIS) data layers prepared in ArcGIS. The Net Logo assisted to incorporate and simulate driving forces that affect location decision-making by households and the growth of informal structures. A preliminary survey was conducted to obtain household location decision preferences. The study established that the development of unplanned structures and eventual settlement has been a function of land price, proximity to economic Centre’s, household economic potential, and the location decision-making patterns of households. This exploratory study also found that the majority of spontaneous development took place in areas liable to floods suggesting that some structures fall outside the required building regulations. The application of the proposed model indicates its potential to improve urban planning policies and decision-making processes in emerging cities of developing countries. Also, the result of the simulation suggests potential preferential location for residential development. The research justifies an approach in the area of simulating urban dynamics with agent-based models given the inclusion of empirical data layers.