Beyond The 100% Assumption: Adjusting OnSSET To Reflect Grid Reliability Costs

OnSSET (Open-Source Spatial Electrification Tool) is a GIS-based tool developed to support electrification planning and decision-making for the achievement of energy access goals in currently unserved locations. OnSSET chooses the “optimal” technology based on a Levelized Cost of Electricity (LCOE) comparison. While the reliability of power supply is explicitly reflected in the sizing of off-grid systems (e.g., mini-grids), the model does not account for (potentially) unreliable supply in new grid connections. 

To address this, we have introduced two variations of the standard LCOE formula that serve as proxies for the cost of non-supplied electricity and test the model's reaction on a case study in Uganda. We observed that the:

1. Cost of Non-Served Energy (CNSE) option - which applied a parametric economic penalty for outages based on reliability indices (e.g., SAIDI) - caused minor technology shifts from grid connections towards solar home systems (SHS); it had only a limited impact on total investment estimates as it functions primarily as a shadow price.
2. Diesel Backup option - which models the explicit CAPEX and OPEX of backup generators required to cover supply gaps in each settlement - resulted in more evident shifts from grid extension to off-grid solutions and also increased the total estimated investment of the least-cost electrification scenario.

The analysis confirms that incorporating reliability metrics may alter the “least-cost” electrification pathway. When the (un)reliability-incurred costs are factored into the modelling process - whether through economic penalties or backup infrastructure costs - decentralized solutions could become a more suitable electrification option in areas with an unstable centralized grid network.

While both methods have limitations (e.g., CNSE is value-laden; diesel backups rely on broad customer assumptions), the new modification represents a long-awaited, critical step towards a more reality-reflective version of OnSSET. We invite planners, researchers, and policymakers to join us in strengthening the approach so that least-cost modelling continues to evolve and better reflect real-world technical constraints, policy environments, and country priorities.