Full Text Available
Note: Clicking the button above will open the full text document at the original institutional repository in a new window.
Operational scheduling and investigative planning for electric public transport in South Africa
Thesis (PhD)--Stellenbosch University, 2025.
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Published: |
Stellenbosch : Stellenbosch University
2026
|
| Subjects: | |
| Tags: |
No Tags, Be the first to tag this record!
|
| _version_ | 1867613911342120960 |
|---|---|
| access_status_str | Open Access |
| author | Wüst, Jacques |
| author2 | Booysen, M. J (Thinus) |
| author_browse | Booysen, M. J (Thinus) Wüst, Jacques |
| author_facet | Booysen, M. J (Thinus) Wüst, Jacques |
| author_sort | Wüst, Jacques |
| collection | Thesis |
| dc_rights_str_mv | Stellenbosch University |
| description | Thesis (PhD)--Stellenbosch University, 2025. |
| format | Thesis |
| id | oai:scholar.sun.ac.za:10019.1/134876 |
| institution | Stellenbosch University (South Africa) |
| last_indexed | 2026-06-10T12:43:40.048Z |
| license_str | Other — see source repository |
| provenance_str_mv | Harvested via OAI-PMH from SUNScholar — Stellenbosch University Repository |
| publishDate | 2026 |
| publishDateRange | 2026 |
| publishDateSort | 2026 |
| publisher | Stellenbosch : Stellenbosch University |
| publisherStr | Stellenbosch : Stellenbosch University |
| record_format | dspace |
| source_str | SUNScholar — Stellenbosch University Repository |
| spelling | oai:scholar.sun.ac.za:10019.1/134876 Operational scheduling and investigative planning for electric public transport in South Africa Wüst, Jacques Booysen, M. J (Thinus) Bekker, J. C. (Nelius) Stellenbosch University. Faculty of Engineering. Dept. of Electrical and Electronic Engineering. Sustainable transportation -- South Africa Paratransit services -- South Africa Battery charging stations (Electric vehicles) Transportation -- South Africa -- Planning Thesis (PhD)--Stellenbosch University, 2025. Wust, J. 2025. Operational Scheduling and Investigative Planning for Electric Public Transport in South Africa. Unpublished doctoral dissertation. Stellenbosch: Stellenbosch University [online]. Available: https://scholar.sun.ac.za/items/6c248117-56cd-4ecf-98ac-a11a67d148cb ENGLISH ABSTRACT: Globally, transport systems face mounting challenges of congestion, pollution, and resource constraints, with projections indicating 68% of the world’s population will reside in urban areas by 2050. In Sub-Saharan Africa (SSA), informal paratransit networks constitute the primary mobility solution. However, these systems operate unscheduled and with fragmented ownership structures, resulting in inefficient vehicle utilisation and substantial environmental consequences through emissions from ageing, poorly maintained fleets. While electric vehicles (EVs) present opportunities to address these challenges, their integration into paratransit operations introduces complexities related to range limitations, charging infrastructure, and vehicle scheduling. These issues are more acute in SSA contexts, where rolling power blackouts (load shedding) impact charging infrastructure reliability, capital constraints often prohibit full fleet electrification, passengers are more flexible and depot scenarios are nuanced. Existing scheduling solutions fail to adequately account for these region-specific constraints, creating a need for tailored approaches. This research addressed two principal questions: How can EVs be scheduled efficiently within SSA’s public transport sector to address the region’s diverse operational scenarios? And can scheduling strategies inform planning for the transition to EVs? The thesis starts with a novel heuristic algorithm for EV fleet scheduling. The heuristic algorithm features a unique backtracking mechanism and incorporates a time-edge approach for charging scheduling. Its capabilities include timetabled charger availability (accounting for load shedding), temporal flexibility (allowing minor trip delays to produce feasible schedules), and provisions for multiple depots, mixed fleets and multi-objective optimisation. These capabilities make the proposition of EVs for public transport more appealing by showcasing solutions that are less capital-intensive. The research methodology spans four studies. In the first study, the heuristic algorithm was developed and tested against various operational scenarios. In the second study, a mixed integer linear programming model was developed with a unique mathematical formulation to solve the Electric Vehicle Scheduling Problem exactly, and a generalised metaheuristic objective function was programmed to use in various metaheuristic algorithms. These techniques were compared to evaluate relative performance in terms of processing speed, solution quality, and implementation complexity. The third and fourth studies applied the heuristic algorithm to real-world data to arrive at valuable insights regarding the transformation of paratransit to EVs. First, real-world GPS taxi tracking data from South Africa was used to compare scheduled and non-scheduled approaches for electric taxi operations. Second, the algorithm was utilised to investigate solar photovoltaic and energy storage systems with EV charging infrastructure under various load shedding scenarios. The research yields the following findings. The developed heuristic algorithm demonstrates superior computational efficiency compared to exact and metaheuristic approaches, processing a 420-trip dataset in 37 seconds, while other methods become computationally intractable. For small problem instances (under 50 trips), metaheuristic approaches may offer slight distance savings worth their computational cost, but the heuristic approach provides the best balance of solution quality, feasibility and computational efficiency for large-scale transit operations. When applied to real-world data, the research demonstrates that a scheduled approach to paratransit electrification outperforms existing mobility patterns. In the Stellenbosch case study, the scheduled scenario achieved an electrification ratio of 85% (11 EVs out of 13 vehicles) compared to just 36% (five EVs out of 14 vehicles) for the non-scheduled approach. Despite requiring a similar total fleet size, the scheduled approach required only five chargers to service 11 EVs, while the non-scheduled approach required four chargers for five EVs. The scheduled approach resulted in a 94% reduction in local CO2 emissions when incorporating renewable energy, compared to 58% for the non-scheduled approach. The investigation of solar-battery charging systems under grid constraints reveals that partial fleet electrification yields better financial outcomes for operators investing in such systems. The analysis of various load shedding scenarios demonstrates that renewable energy investments become more valuable in regions with harsher grid constraints, and that the timing of these investments depends on the grid’s reliability and scale of fleet conversion. The study concludes that EVs can be scheduled efficiently within the unique context of SSA’s public transport sector through heuristic optimisation techniques tailored to the region-specific constraints. Furthermore, scheduling strategies inform planning for the transition to EVs by revealing the feasibility of various electrification scenarios, guiding charger infrastructure development, and quantifying the benefits of renewable energy integration. AFRIKAANSE OPSOMMING: Wêreldwyd staar vervoerstelsels toenemende uitdagings van opeenhoping, besoedeling, en hulpbronbeperkings in die gesig, met voorspellings wat aandui dat 68% van die wêreld se bevolking teen 2050 in stedelike gebiede gaan woon. In Sub-Sahara Afrika (SSA) word mobiliteit hoofsaaklik voorsien deur informele paratransitnetwerke. Hierdie stelsels funksioneer egter sonder skedules en met verdeelde strukture van eienaarskap, wat lei tot ondoeltreffende voertuigbenutting en aansienlike gevolge vir die omgewing weens uitlaatgasse van verouderde voertuigvlote wat swak onderhou word. Terwyl elektriese voertuie geleenthede bied om hierdie uitdagings aan te spreek, skep hul integrasie in paratransit kompleksiteite rakende afstandbeperkings, laaistasie infrastruktuur, en voertuigskedulering. Hierdie kwessies is meer akuut in die SSA-konteks, waar kragonderbrekings (beurtkrag) die betroubaarheid van laaistasie infrastruktuur beïnvloed, beperkings in terme van kapitaal dikwels volledige vlootelektrifisering verhoed, passasiers meer buigsaam is en depot scenarios meer uniek is. Bestaande skeduleringsoplossings neem nie hierdie streekspesifieke beperkings voldoende in ag nie, wat ’n behoefte skep vir aangepaste benaderings. Hierdie navorsing beantwoord die volgende twee navorsingsvrae: Hoe kan elektriese voertuie doeltreffend geskeduleer word in SSA se openbare vervoersektor om die streek se diverse operasionele omstandighede aan te spreek? En, kan skeduleringsstrategieë beplanning vir die oorgang na elektriese voertuie ondersteun? Die tesis begin met die aanbieding van ’n nuwe heuristiese algoritme vir elektriese vlootskedulering wat ontwikkel is. Die heuristiese algoritme bevat ’n unieke erugspoormeganisme en inkorporeer ’n tyd-rand-benadering vir laaiskedulering. Die vermoëns sluit in geskeduleerde laaibeskikbaarheid (wat beurtkrag in ag neem), tydsbuigsaamheid (wat klein ritvertragings toelaat om uitvoerbare skedules te skep), en voorsiening vir veelvuldige depots, gemengde vlote en meervoudige-doelwit optimering. Hierdie vermoëns maak elektriese voertuig voorstelle vir openbare vervoer meer aanloklik deur oplossings te vind wat minder kapitaalintensief is. Die navorsingsmetodologie strek oor vier studies. In die eerste studie is die heuristiese algoritme ontwikkel en getoets in verskeie operasionele situasies. In die tweede studie was ’n gemengde heeltallige lineêre programmeringsmodel ontwikkel met ’n unieke wiskundige formulering om die elektriese voertuig-skeduleringsprobleem presies op te los, en ’n veral gemeende metaheuristiese doelfunksie was geprogrammeer om in verskeie metaheuristiese algoritmes te gebruik. Hierdie tegnieke was met mekaar vergelyk om relatiewe verrigting te evalueer in terme van verwerkingspoed, oplossingsgehalte, en implementeringskompleksiteit. Die derde en vierde studies pas die heuristiese algoritme toe op werklike data om waardevolle insigte te verkry rakende die transformasie van paratransit na elektriese voertuie. Eerstens word GPS taxi spoordata van Suid-Afrika gebruik om geskeduleerde en nie-geskeduleerde benaderings vir elektriese taxi-operasies te vergelyk. Tweedens word die algoritme gebruik om sonkrag-fotovoltaïese en energiestoorstelsels met laai-infrastruktuur onder verskeie beurtkragscenario’s te ondersoek. Die navorsing lewer die volgende bevindings. Die ontwikkelde heuristiese algoritme toon beter berekeningsdoeltreffendheid as presiese en metaheuristiese benaderings. Dit verwerk ’n datastel van 420 ritte in 37 sekondes, terwyl ander metodes sukkel. Vir klein probleme (minder as 50 ritte) kan metaheuristiese benaderings geringe afstandbesparings bied wat hul berekeningskoste werd is, maar die heuristiese benadering bied die beste balans tussen oplossingsgehalte en berekeningsdoeltreffendheid vir grootskaalse vervoerstelsels. Wanneer dit op werklike data toegepas word, toon die navorsing dat ’n geskeduleerde benadering tot paratransit-elektrifisering beter doen as bestaande mobiliteitspatrone. In die Stellenbosch-gevallestudie het die geskeduleerde scenario ’n elektrifiseringsverhouding van 85% (11 elektriese voertuie uit 13 voertuie) behaal, in vergelyking met slegs 36% (vyf elektriese voertuie uit 14 voertuie) vir die nie-geskeduleerde benadering. Ten spyte van ’n soortgelyke vlootgrootte, het die geskeduleerde benadering slegs vyf laaiers benodig om 11 elektriese voertuie te bedien, terwyl die nie-geskeduleerde benadering vier laaiers vir vyf elektriese voertuie benodig het. Die geskeduleerde benadering het gelei tot ’n 94% vermindering in plaaslike CO2-uitlaatgasse wanneer hernubare energie geïnkorporeer word, in vergelyking met 58% vir die nie-geskeduleerde benadering. Die ondersoek van sonkrag-battery-laaistelsels onder kragnetwerkbeperkings toon dat gedeeltelike vlootelektrifisering beter finansiële uitkomste lewer. Die ondersoek van verskeie beurtkragscenario’s toon dat beleggings in hernubare energie meer waardevol word in streke met strenger kragnetwerkbeperkings, en dat die tydsberekening van hierdie beleggings afhang van die betroubaarheid van die kragnetwerk en die skaal van vlootomskakeling. Die studie kom tot die gevolgtrekking dat elektriese voertuie doeltreffend geskeduleer kan word binne die unieke konteks van SSA se openbare vervoersektor deur heuristiese optimeringstegnieke wat aangepas is vir die streek se spesifieke beperkings. Verder kan skeduleringsstrategieë beplanning vir die oorgang na elektriese voertuie ondersteun deur die haalbaarheid van verskeie elektrifiseringsscenario’s te openbaar, leiding te gee vir die ontwikkeling van laaier-infrastruktuur, en die voordele van hernubare energie-integrasie te kwantifiseer. Doctoral 2026-01-13T12:46:52Z 2026-01-13T12:46:52Z 2025-12 Thesis https://scholar.sun.ac.za/handle/10019.1/134876 Stellenbosch University xvi, 177 pages : illustrations application/pdf Stellenbosch : Stellenbosch University |
| spellingShingle | Sustainable transportation -- South Africa Paratransit services -- South Africa Battery charging stations (Electric vehicles) Transportation -- South Africa -- Planning Wüst, Jacques Operational scheduling and investigative planning for electric public transport in South Africa |
| title | Operational scheduling and investigative planning for electric public transport in South Africa |
| title_full | Operational scheduling and investigative planning for electric public transport in South Africa |
| title_fullStr | Operational scheduling and investigative planning for electric public transport in South Africa |
| title_full_unstemmed | Operational scheduling and investigative planning for electric public transport in South Africa |
| title_short | Operational scheduling and investigative planning for electric public transport in South Africa |
| title_sort | operational scheduling and investigative planning for electric public transport in south africa |
| topic | Sustainable transportation -- South Africa Paratransit services -- South Africa Battery charging stations (Electric vehicles) Transportation -- South Africa -- Planning |
| url | https://scholar.sun.ac.za/handle/10019.1/134876 |
| work_keys_str_mv | AT wustjacques operationalschedulingandinvestigativeplanningforelectricpublictransportinsouthafrica |