The Department of Statistical Sciences and the Clinton Health Access Initiative have come together to provide an exciting opportunity for Masters students through an fellowship programme. The programme provides an opportunity for students in their dissertation year to analyse government data to answer a policy-relevant question enabling them to receive work experience that is aligned with their dissertation. MASHA facilitated the establishment of this programme and is proud to present the STA-CHAI fellowship students below.

 

2024 Cohort 

Student Grace Carmichael

​Grace Carmichael
Home Country: South Africa 
E-mail: crmgra004@myuct.ac.za / grace.carmichael@uct.ac.za

Education

  • BSc in Applied Statistics and Genetics at UCT.
  • Honours in Statistical Sciences at UCT. 
Student Suzana Maganga_5

Susana Maganga
Home Country: Tanzania
E-mail: MGNSUS002@myuct.ac.za

Education

  • BAgric Agribusiness Management at Stellenbosch University.
  • BCom (Hons) Statistics at Stellenbosch University.

Student Ropafadzo Chimuti_5

Ropafadzo Chimuti
Home Country: Zimbabwe 
E-mail: CHMROP002@myuct.ac.za

Education

  • BSc in Geomatics at UCT.

Thesis Title:  Monitoring Fleet Vehicle Usage using GPS Data
Abstract: Efficient fleet management is essential for enhancing resource allocation and operational performance, particularly in demand-driven contexts. This study examines the utilisation of the Eastern Cape Department of Health’s white fleet vehicles, with three objectives: (1) assessing utilisation at site and site-group levels, (2) distinguishing typical from atypical trips, and (3) identifying over- and underutilised vehicles. Drawing on GPS data from 498,601 trips over one year, the analysis applied metrics such as trip counts, inactivity ratios, and simultaneous use ratios, alongside anomaly detection methods (Isolation Forest and DBSCAN), to flag 58 vehicles with irregular usage patterns.

Findings highlight substantial disparities in utilisation: hospitals generally demonstrate higher efficiency, while administrative offices and community health centres show both underutilisation and heavy reliance on a small number of vehicles.

Recommendations include redistributing vehicles to balance demand, adopting dynamic scheduling, investigating atypical trip patterns, and implementing predictive maintenance strategies. The study offers not only actionable insights for the Eastern Cape but also a scalable framework for enhancing fleet management in public health systems more broadly.