ABSTRACT
Due to the dynamic climatic conditions of Pakistan, the planning and management of electricity have become challenging tasks. All planning foundations build on accurate electricity demand forecasts, and demand overestimation may result in non-use payments for generation capacity. This study provides a detailed examination of the Indicative Generation Capacity Expansion Plan (IGCEP) demand forecast model. The GCEP is a pivotal snapshot of power planning developed by the National Transmission and Dispatch Company (NTDC) and approved by the National Electric Power Regulatory Authority (NEPRA). The IGCEP, integral to future electricity generation planning, relies on the accuracy of electricity demand projections. However, since 2018, the IGCEP forecasting model has exhibited overfitting issues, employing an annual data approach that overlooks seasonal variations. To address the challenge of overfitting, this study examines the factors that contribute to overfitting in aggregated data forecasts and finds that data aggregation influences forecast accuracy. A key aspect of this study is the exploration of electricity sales and demand as distinct entities. While the NTDC and other stakeholders in the Ministry of Energy assume that electricity sales are equivalent to demand. This research establishes that electricity planning on demand and sales are significantly different phenomena.

The findings not only rectify the overfitting issues in the IGCEP energy demand forecasting model but also offer a deep understanding of the factors influencing electricity demand in Pakistan. The study suggested a paradigm shift by moving the electricity demand forecast model from a purely economic perspective to a meteorological one. Ultimately, this shift towards a meteorological perspective provides a more accurate foundation for planning future electricity demand in Pakistan, ensuring the sustainability and reliability of power planning