Assessing Community Awareness of Water Infrastructure Encroachment Risks in Makause Informal Settlement, City of Ekurhuleni
Mpondomise Nkosinathi Ndawo *
Management College of Southern Africa, MANCOSA, Research Directorate, 26 Samora Machel Street, Durban 4001, South Africa.
Stephen Loh Tangwe
Central University of Technology, Resources and Operations Division, Bloemfontein 9301, Free State Province, South Africa.
*Author to whom correspondence should be addressed.
Abstract
The management of the bulk water infrastructure is a critical aspect of urban resilience, particularly as cities expand and informal settlements increasingly encroach upon essential services. Encroachment presents significant threats to water supply systems, elevating the risks of infrastructural damage, contamination, and service disruptions. This study examines the risk of informal settlement encroachment on critical water infrastructure in the Makause informal settlement. It aims to identify the key factors influencing encroachment and to develop predictive models that support proactive, community-based infrastructure protection. A mixed quantitative–computational approach was employed, using survey data from 105 residents. Descriptive statistics and one-way ANOVA were applied to evaluate differences across categorical responses (“Yes”, “No”, “Unsure”). The ReliefF algorithm was used to rank variable importance in predicting encroachment risk. Key predictors were then used to train, validate, and test an artificial neural network (ANN) model to assess its suitability for risk forecasting. The ANN achieved high predictive accuracy, with correlation coefficients exceeding 0.95 and low mean squared error values across all modelling phases. ANOVA results confirmed statistically significant differences among selected variables. ReliefF identified community awareness, settlement proximity, and resource access as the most influential predictors. Model validation showed strong agreement between predicted and actual outcomes (p > 0.900), confirming robustness and reliability. This study proposes a novel, data-driven framework that integrates machine learning and statistical analysis for infrastructure risk assessment in informal settlements. It demonstrates how community-generated data can be combined with computational techniques to strengthen urban infrastructure management. The framework offers municipalities and water utilities a practical tool for engaging communities, prioritising interventions, and improving protection of critical infrastructure in rapidly urbanising environments. Results are based on a single case study in Makause and may reflect self-reporting bias. A broader application would require additional case studies and expanded datasets.
Keywords: Encroachment, informal settlements, water infrastructure, ReliefF, artificial neural network, infrastructure risk