{"id":{"repo_id":"stellenbosch","oai_identifier":"oai:scholar.sun.ac.za:10019.1/136252"},"canonical_url":"https://search.dev.ndltd.org/etd/stellenbosch/oai:scholar.sun.ac.za:10019.1/136252","repository":{"repo_id":"stellenbosch","name":"Stellenbosch University","base_url":"https://scholar.sun.ac.za/server/oai/request"},"display":{"title":"A Sociotechnical Systems Approach to Road Safety in a Low- and Medium- Income Country: Gauteng Province South Africa","abstract":"This PhD thesis investigates road fatalities in Gauteng Province, South Africa, from 2015 to 2020. Using a Sociotechnical Systems framework, the study examines how social and technical elements interact to influence road safety outcomes. Drawing on data analytics and mathematical modelling, the study transitions from traditional statistical methods: Poisson Regression, Multiple Linear Regression and Negative Binomial Regression alongside advanced machine learning algorithms like Random Forest. Furthermore, a novel Spatially Aware Poisson Random Forest ensemble model is developed to capture complex spatial dependencies within the data. Key findings reveal that while vehicle population growth (5.6% annually) correlates positively with fatalities (IRR=1.087), older vehicles (15–20 years) and high-mobility roads (Classes 1–3) amplify risks, moderated by spatial heterogeneity in townships and intersections. Social determinants, such as unemployment (β=0.000957, p=0.092) and holidays (8–13% of annual fatalities), exacerbate vulnerabilities, particularly for pedestrians (46–49% of deaths) and youth (0–14 years: 19.2%). Temporal peaks occur on weekends (61.4%) and winter nights, underscoring visibility and behavioural factors. The spatially aware model achieves superior performance (R²=0.958, MAE=0.031), highlighting intersection proximity and time-of-day as dominant predictors. Through spatiotemporal hotspot analysis (Moran's I=0.2684), the thesis identifies high-risk locales like Soweto and Golden Highway, advocating integrated interventions: targeted infrastructure upgrades, speed management, pedestrian protections, and socioeconomic-aligned policies. This framework advances Sociotechnical road safety paradigms, offering predictive tools for equitable, context-specific mitigation in middle-income settings.","abstract_html":"This PhD thesis investigates road fatalities in Gauteng Province, South Africa, from 2015 to 2020. Using a Sociotechnical Systems framework, the study examines how social and technical elements interact to influence road safety outcomes. Drawing on data analytics and mathematical modelling, the study transitions from traditional statistical methods: Poisson Regression, Multiple Linear Regression and Negative Binomial Regression alongside advanced machine learning algorithms like Random Forest. Furthermore, a novel Spatially Aware Poisson Random Forest ensemble model is developed to capture complex spatial dependencies within the data. Key findings reveal that while vehicle population growth (5.6% annually) correlates positively with fatalities (IRR=1.087), older vehicles (15–20 years) and high-mobility roads (Classes 1–3) amplify risks, moderated by spatial heterogeneity in townships and intersections. Social determinants, such as unemployment (β=0.000957, p=0.092) and holidays (8–13% of annual fatalities), exacerbate vulnerabilities, particularly for pedestrians (46–49% of deaths) and youth (0–14 years: 19.2%). Temporal peaks occur on weekends (61.4%) and winter nights, underscoring visibility and behavioural factors. The spatially aware model achieves superior performance (R²=0.958, MAE=0.031), highlighting intersection proximity and time-of-day as dominant predictors. Through spatiotemporal hotspot analysis (Moran's I=0.2684), the thesis identifies high-risk locales like Soweto and Golden Highway, advocating integrated interventions: targeted infrastructure upgrades, speed management, pedestrian protections, and socioeconomic-aligned policies. This framework advances Sociotechnical road safety paradigms, offering predictive tools for equitable, context-specific mitigation in middle-income settings.","abstract_has_math":false,"creators":["Masuku, Freeman Nkululeko"],"institution":"Stellenbosch : Stellenbosch University","degree_name":null,"degree_level":null,"degree_discipline":null,"degree_department":null,"school":null,"contributors":[],"advisors":["Sinclair, Marion"],"committee_chairs":[],"committee_members":[],"year":2026,"date_issued":"2026-03","date_published":"2026-03","updated_at":"2026-07-24T04:40:06Z","subjects":[],"languages":["en"],"rights":[],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"https://scholar.sun.ac.za/handle/10019.1/136252","outbound_label":"Repository record","outbound_source":"dc:identifier.uri"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Sinclair, Marion"]},{"key":"dc:contributor.other","label":"Dc Contributor Other","values":["Stellenbosch University. Faculty of Engineering. Dept. of Civil Engineering."]},{"key":"dc:creator","label":"Author","values":["Masuku, Freeman Nkululeko"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2026-04-29T12:40:03Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2026-04-29T12:40:03Z"]},{"key":"dc:date.issued","label":"Date","values":["2026-03"]},{"key":"dc:publisher","label":"Institution","values":["Stellenbosch : Stellenbosch University"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language.iso","label":"Language (ISO)","values":["en"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://scholar.sun.ac.za/handle/10019.1/136252"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["Thesis (PhD)--Stellenbosch University, 2026.","Masuku, F. N. 2026. A Sociotechnical Systems Approach to Road Safety in a Low- and Medium- Income Country: Gauteng Province South Africa. Unpublished doctoral dissertation. Stellenbosch: Stellenbosch University [online]. Available: https://scholar.sun.ac.za/items/35e2a838-3702-482e-bab6-3edb9963fa6b"]},{"key":"dc:description.abstract","label":"Abstract","values":["This PhD thesis investigates road fatalities in Gauteng Province, South Africa, from 2015 to 2020. Using a Sociotechnical Systems framework, the study examines how social and technical elements interact to influence road safety outcomes. Drawing on data analytics and mathematical modelling, the study transitions from traditional statistical methods: Poisson Regression, Multiple Linear Regression and Negative Binomial Regression alongside advanced machine learning algorithms like Random Forest. Furthermore, a novel Spatially Aware Poisson Random Forest ensemble model is developed to capture complex spatial dependencies within the data. Key findings reveal that while vehicle population growth (5.6% annually) correlates positively with fatalities (IRR=1.087), older vehicles (15–20 years) and high-mobility roads (Classes 1–3) amplify risks, moderated by spatial heterogeneity in townships and intersections. Social determinants, such as unemployment (β=0.000957, p=0.092) and holidays (8–13% of annual fatalities), exacerbate vulnerabilities, particularly for pedestrians (46–49% of deaths) and youth (0–14 years: 19.2%). Temporal peaks occur on weekends (61.4%) and winter nights, underscoring visibility and behavioural factors. The spatially aware model achieves superior performance (R²=0.958, MAE=0.031), highlighting intersection proximity and time-of-day as dominant predictors. Through spatiotemporal hotspot analysis (Moran's I=0.2684), the thesis identifies high-risk locales like Soweto and Golden Highway, advocating integrated interventions: targeted infrastructure upgrades, speed management, pedestrian protections, and socioeconomic-aligned policies. This framework advances Sociotechnical road safety paradigms, offering predictive tools for equitable, context-specific mitigation in middle-income settings."]},{"key":"dc:title","label":"Title","values":["A Sociotechnical Systems Approach to Road Safety in a Low- and Medium- Income Country: Gauteng Province South Africa"]}]}],"canonical_facts":{"dc:contributor.advisor":["Sinclair, Marion"],"dc:contributor.other":["Stellenbosch University. Faculty of Engineering. Dept. of Civil Engineering."],"dc:creator":["Masuku, Freeman Nkululeko"],"dc:date.accessioned":["2026-04-29T12:40:03Z"],"dc:date.available":["2026-04-29T12:40:03Z"],"dc:date.issued":["2026-03"],"dc:description":["Thesis (PhD)--Stellenbosch University, 2026.","Masuku, F. N. 2026. A Sociotechnical Systems Approach to Road Safety in a Low- and Medium- Income Country: Gauteng Province South Africa. Unpublished doctoral dissertation. Stellenbosch: Stellenbosch University [online]. Available: https://scholar.sun.ac.za/items/35e2a838-3702-482e-bab6-3edb9963fa6b"],"dc:description.abstract":["This PhD thesis investigates road fatalities in Gauteng Province, South Africa, from 2015 to 2020. Using a Sociotechnical Systems framework, the study examines how social and technical elements interact to influence road safety outcomes. Drawing on data analytics and mathematical modelling, the study transitions from traditional statistical methods: Poisson Regression, Multiple Linear Regression and Negative Binomial Regression alongside advanced machine learning algorithms like Random Forest. Furthermore, a novel Spatially Aware Poisson Random Forest ensemble model is developed to capture complex spatial dependencies within the data. Key findings reveal that while vehicle population growth (5.6% annually) correlates positively with fatalities (IRR=1.087), older vehicles (15–20 years) and high-mobility roads (Classes 1–3) amplify risks, moderated by spatial heterogeneity in townships and intersections. Social determinants, such as unemployment (β=0.000957, p=0.092) and holidays (8–13% of annual fatalities), exacerbate vulnerabilities, particularly for pedestrians (46–49% of deaths) and youth (0–14 years: 19.2%). Temporal peaks occur on weekends (61.4%) and winter nights, underscoring visibility and behavioural factors. The spatially aware model achieves superior performance (R²=0.958, MAE=0.031), highlighting intersection proximity and time-of-day as dominant predictors. Through spatiotemporal hotspot analysis (Moran's I=0.2684), the thesis identifies high-risk locales like Soweto and Golden Highway, advocating integrated interventions: targeted infrastructure upgrades, speed management, pedestrian protections, and socioeconomic-aligned policies. This framework advances Sociotechnical road safety paradigms, offering predictive tools for equitable, context-specific mitigation in middle-income settings."],"dc:identifier.uri":["https://scholar.sun.ac.za/handle/10019.1/136252"],"dc:language.iso":["en"],"dc:publisher":["Stellenbosch : Stellenbosch University"],"dc:title":["A Sociotechnical Systems Approach to Road Safety in a Low- and Medium- Income Country: Gauteng Province South Africa"],"dc:type":["Thesis"]},"updated_at":"2026-07-24T04:40:06Z"}