Chronic inefficiencies in power-distribution infrastructure contribute significantly to electricity losses and supply instability. There is a pressing need for robust, field-based methodologies to quantify the impact of infrastructure upgrades within complex, real-world grid environments. This study aimed to develop and apply a quasi-experimental design to empirically measure efficiency gains from the deployment of advanced distribution transformers and line-voltage regulators within a national utility network. A difference-in-differences framework was employed, comparing treated and control feeder groups before and after equipment installation. Technical losses were the primary outcome. The core statistical model was Y₈ₓ = ₀ + ₁ Treatᵢ + ₂ Postₜ + (Treatᵢ Postₜ) + ₈ₓ, with inference based on cluster-robust standard errors at the substation level. The intervention produced a statistically significant reduction in average technical losses of 4. 7 percentage points (95% CI: 3. 1 to 6. 3). This effect was robust to seasonal adjustments and variations in feeder loading patterns. The quasi-experimental design provided credible, field-based evidence of substantial efficiency improvements from targeted equipment upgrades, validating the methodological approach for infrastructure evaluation. Utilities should adopt quasi-experimental designs for pilot programmes to generate rigorous evidence for capital investment decisions. Regulatory frameworks should incentivise investments proven to reduce technical losses. quasi-experimental design, distribution losses, difference-in-differences, infrastructure efficiency, grid modernisation This paper provides a novel application of a quasi-experimental, causal-inference framework to evaluate engineering interventions in a live power-distribution network, generating robust evidence for asset-management policy.
Nkosi et al. (Tue,) studied this question.