The rapid adoption of electric vehicles (EVs) in Kenya and other emerging markets is introducing spatially concentrated loads onto radial distribution networks not originally designed for them. Although Kenya's electricity mix is rich in geothermal, hydro, wind and solar resources, uncoordinated placement of charging infrastructure deepens voltage drops, raises feeder losses and erodes the operating margin of already weak buses. This study formulates EV charging-station placement and sizing as a constrained, nonlinear, voltage-stability-aware distribution-planning problem and solves it using the Artificial Bee Colony (ABC) algorithm.

The IEEE 33-bus radial distribution network is modelled in MATLAB/Simulink, with a backward-forward sweep load-flow routine embedded inside the optimization loop. EV charging demand is superimposed at candidate buses under low, medium and high penetration scenarios reflecting Nairobi's emerging urban charging landscape. A baseline assessment quantifies how uncoordinated placement degrades the minimum-bus voltage, active-power losses and selected voltage-stability indices. The ABC algorithm then searches for siting and sizing solutions that minimize a weighted fitness combining normalized real-power loss and voltage deviation, subject to power-flow, bus-voltage and charger-placement constraints, with penalty handling for infeasible solutions. Repeatability is assessed across multiple runs given the stochastic nature of the metaheuristic.

Relative to uncoordinated deployment, the ABC-optimized placement raises the minimum-bus voltage, reduces total active-power losses, improves voltage-stability margins and increases the feeder's EV hosting capacity across all penetration levels, with stable convergence under tuned ABC parameters. Sensitivity studies on station number, charger rating and fitness weights confirm robustness under alternative planning assumptions.

The work demonstrates that ABC is a suitable metaheuristic for voltage-stability-aware charging-infrastructure planning and that network-aware siting delivers measurable technical gains without additional feeder reinforcement. The central conclusion is that charging-station placement in emerging markets should be treated as a distribution-planning problem grounded in power-flow performance, not solely as a transport-access or economic one. Findings are relevant to utilities, regulators and electric-mobility planners in Kenya and comparable Sub-Saharan African contexts scaling EV charging on unevenly strong feeders