For decades, Pakistan's electricity supply has fluctuated between surplus and deficit. Recently, the country has achieved surplus generation capacity, leading to significant circular debt due to high capacity payments. Additionally, daily and seasonal demand variations contribute to this surplus during off-peak times, which has drastically increased the Weighted Average Cost of Generation (WACG). To address this, there is a need to find suitable, non-seasonal, flexible loads that do not increase daily load peaks. A dynamic, temporally variable tariff model is essential to encourage generation capacity utilization during off-peak hours. This approach benefits both energy consumers and the power sector: consumers get better rates, while the power sector improves its load factor by using excess capacity during off-peak times. To address this challenge, we developed a dynamic, dashboard-based Time-of-Use (ToU) pricing tool. The tool adjusts tariff rates, making them higher during peak hours to incentivize off-peak electricity usage. Importantly, the tool ensures tariffs remain above the basket price of electricity, preventing any additional burden on the national budget. The tool's effectiveness was demonstrated through various scenarios. For example, shifting just 5% of peak load to off-peak hours in August could save the power sector 17.18 million rupees. Similarly, shifting power plant operations to off-peak hours could generate 93.5 million rupees per month in emissions reduction value for October. CO2 emissions would also decrease from 5413.08 metric tons to 5044.30 metric tons with a 5% demand shift, aligning with the UN Sustainable Development Goals. This load-shifting method could help reduce circular debt and carbon emissions. By lowering the WACG, the power sector can cut capacity payments, reduce debt, and provide revenue to generation companies. Our ToU pricing tool offers affordable off-peak rates, benefiting consumers, power generation entities, and the government. Finally, we recommend strategies to maximize the benefits of this mechanism.

Short-Term Load Forecasting (STLF) is a critical tool for managing energy markets. STLF utilizes various inputs to generate accurate load forecasts for specific regions. Several techniques are available for STLF, including statistical methods like multiple regression, exponential smoothing, iterative reweighted least squares, adaptive load forecasting, and stochastic time series. Additionally, artificial intelligence-based techniques such as fuzzy logic, genetic programming, deep neural networks, random forests, and support vector machines have been employed, along with knowledge-based methods like case-based reasoning. However, no single technique is universally suitable for every region's forecasting needs. Therefore, a hybrid model that combines multiple techniques is necessary to achieve accurate predictions. The tool developed by LEI uses automated inputs such as historical demand data and weather forecasts. Manual inputs can also be added for better accuracy, including significant events like Ramadan, elections, holidays, and schedules for offices and schools. The tool is customizable, allowing it to be tailored to the specific needs of Distribution Companies (DISCOs) or specific areas within them. Currently, the tool is being tested on offline historical load data. Based on this data alone, it provides a Mean Absolute Percentage Error (MAPE) of 7% to 1%, depending on the season. With additional inputs, the tool is expected to reduce its daily MAPE to 0.5% to 3%. LUMS' Energy Informatics Group (EIG) has deployed a team of developers and data scientists for this project.

This electricity outlook report presents the findings from the independent dispatch model developed by the LUMS Energy Institute (LEI) for Pakistan’s power system. It critically examines the Indicative Generation Capacity Expansion Plan (IGCEP) 2021 by NTDC, offering a thorough analysis of the business-as-usual trajectory outlined in IGCEP 2021. The aim is to provide readers with a clear understanding of Pakistan's power system. The report covers nine financial years, from 2021-22 (FY22) to 2029-30 (FY30), aligning with the timeline in IGCEP 2021. The modeling process utilizes the in-house LUMS Power Dispatch Model (LPDM), which simulates the NTDC system to replicate the IGCEP 2021 scenario. Inputs and assumptions are based on IGCEP 2021, with updates from the latest available data from relevant organizations. The results include analyses of capacity and energy balance, fuel-based dispatch, capacity and energy payments, the impact on power purchase prices, and projected fuel quantities. Additionally, nine alternative scenarios have been developed and simulated using LPDM, exploring the effects of changes in key variables. This report provides a summary of key takeaways and conclusions from the IGCEP 2021 scenario, as well as insights from the alternative scenarios. However, this summary only scratches the surface. Readers are encouraged to review the full report for a comprehensive understanding, especially when considering the alternative scenarios.

The LUMS Energy Institute (LEI) conducted the PRECON (Pakistan Residential Energy Consumption Dataset) project. This initiative is the first in Pakistan to collect extensive, appliance-level electricity consumption data from residential households. The project, conducted in 42 selected households in Lahore, provides highly detailed one-minute granularity data, especially for high-energy devices like air conditioners. This unique dataset is crucial for understanding residential energy demand and shaping future energy optimization efforts, including distributed solar energy adoption. The PRECON dataset not only benefits Pakistan but also has potential implications for energy policies in neighboring regions, such as Northern India, with similar climatic conditions. By offering this detailed and representative dataset, LEI aims to support informed decision-making in energy management, policy formulation, and sustainable energy solutions in Pakistan and the broader South Asian region.

This project aimed to improve energy efficiency and conservation in government buildings across Pakistani municipalities. It involved installing energy monitoring systems that provided high-resolution data on energy use with one-minute precision. This data was crucial for conducting detailed energy audits and analysing consumption patterns to identify inefficiencies and develop strategies for improvement. The project also included extensive training for municipal staff on energy management systems (EnMS), demand-side management (DSM) techniques, and practical energy-saving measures. The goal was to equip staff with the skills to implement effective energy conservation practices. Additionally, the project focused on creating policy recommendations to enhance energy management in government buildings. By combining data analysis, training, and policy development, the project aimed to improve energy practices, promote sustainable use, and reduce overall consumption in the public sector.