A Methodology for Bidding Zone Configuration in the Korean Electricity Market Using Public Power Grid Investment Plans

Abstract

Uniform electricity pricing systems, such as Korea's current SMP market, face structural limitations as they fail to reflect transmission congestion costs in price signals. This can exacerbate “spatial mismatches” between generation facilities (often in non-capital areas) and demand centers (in the capital area), leading to significant inefficiencies such as growing redispatch costs and increasing renewable energy curtailment. As countries with uniform pricing consider transitioning to zonal systems, they face a critical gap: the lack of a concrete and applicable methodology for configuring bidding zones. Existing international studies are often difficult to apply, as they presuppose data (e.g., nodal LMPs, detailed redispatch records) that are unavailable in single-price markets. This study proposes a novel, data-driven methodology to overcome this data limitation. Instead of non-existent market data, it utilizes publicly available Long-term Transmission and Substation Facility Plan (LTSFP) data. Based on the premise that future structural congestion is reflected in large-scale government investment plans designed to alleviate it, this study develops two quantitative indices: the Investment Concentration Index (ICI), to measure the spatial density of future investments, and the Bottleneck Persistence Score (BPS), to quantify the chronic nature of bottlenecks across successive plans. Applying this methodology to the Korean power grid, the study identifies the West corridor (Honam–Chungcheong) and the East corridor (Yeongdong–Capital) as the core structural seams. Based on these findings, a 5-zone configuration (Capital, Chungcheong, Yeongdong, Honam, Yeongnam) reflecting these physical constraints is proposed. As a case study, PLEXOS-based zonal market simulations using a flow-based DC-OPF are conducted to compare the proposed 5-zone configuration with the government’s 2-zone (Capital vs. Non-Capital) proposal under identical demand, fuel price, and network assumptions. The simulations show that the 5-zone design generates more distinct and granular zonal price patterns that are closely aligned with the underlying structural congestion and reflect the timing and location of renewable generation surpluses. Although the model does not explicitly simulate ex-post redispatch, these results illustrate how internalizing structural congestion at bidding-zone borders and reflecting it in zonal prices can strengthen locational signals and potentially reduce the need for corrective actions in actual system operation. This study holds significant policy and academic value as it is the first to present an objective, practical, and forward-looking methodology for bidding zone configuration based on public planning data rather than unavailable market data.