Development of a Cellular Automata-based model approach for sustainable planning of affordable housing projects: an application case study in Algiers

Abstract

Being considered as a basic necessity and a key element in the development of sustainable communities, housing is a major concern for the Algerian government. Algiers is facing for the last years a challenging housing shortage and as a way to overcome this problem, many housing projects have been launched. However, the pressing need to address this crisis has disregarded what the pattern of landscape will be, how the existing infrastructures will accommodate with such housing projects and how they will impact on human well-being. This research aims to advance the challenges of planning for sustainability by proposing a methodological approach in a context of high lack of data to support decision-makers in the elaboration of affordable housing projects. The main objective is to trade off urban growth with residential satisfaction and the preservation of natural resources. We developed two Cellular Automata (CA) based residential development scenarios to identify suitable locations for future affordable housing projects: Urban densification scenario (UD) and Constrained urban sprawl scenario (CUS). Both scenarios are based on indicators of residential preferences and measures taken in order to counter the negative effects of urbanization. Results reveal the low capacity of Algiers to meet the housing shortage according to the conditions set for each scenario. The scenarios were evaluated by quantifying their spatial patterns using a preselected set of six class-level landscape metrics. Results show a combination of aggregated and dispersed patterns growth for both scenarios meeting trade-offs among the advantages and the challenges of urban densification and urban sprawl. Then, Standard deviation and regression analysis were conducted to assess the accuracy of CA simulation and the evaluation of pattern changes in the simulated scenarios. The resulting values indicate the good performance of CA and confirm its effectiveness in predicting the future locations of housing projects.