Skip to Main Content

Probabilistic Reliability Modeling

Brattle has experience evaluating the drivers of resource adequacy risks across many North American markets. For our clients, we have developed custom in-house tools and have also utilized commercially-available software to simulate standard reliability metrics including Expected Unserved Energy (EUE), Loss of Load Expectation (LOLE), and Loss of Load Hours (LOLH) to help our clients make decisions regarding appropriate planning reserve margins needed to achieve reliability standards.

These reliability modeling tools probabilistically simulate the major drivers of resource adequacy risk, including generation outages, weather and other load uncertainty, intertie availability, intermittency of wind and solar generation, and availability of demand-side resources. Our simulations realistically account for market operating procedures, including hourly generation dispatch, import-export scheduling, ancillary services, demand response, and individual emergency procedures. In addition, we often simulate how planning reserve margin decisions affect underlying system economics and costs, including annual production costs, customer costs, market prices, load shed costs, and generator energy margins. A unique feature of our models is the ability to capture flexibility-driven versus reliability-driven reliability events as well as needs for flexible resource requirements, intra-hour price volatility, and expected curtailment levels.

To correctly evaluate the uncertainties we simulate them under the most likely conditions (“Base Case”), a number of sensitivity cases, and a range of planning reserve margin levels. The results for each single case and simulated reserve margin level reflect the probability-weighted outcomes for thousands of full annual simulations.


Below is a list of representative engagements for our Probabilistic Reliability Modeling practice.

Evaluation of investment incentives and resource adequacy in ERCOT

For ERCOT, Brattle experts characterized the factors influencing generation investment decisions, analyzed the energy market’s ability to support investment and resource adequacy at the target level based on probabilistic simulation analyses, and evaluated options to enhance long-term resource adequacy while maintaining market efficiency. We interviewed stakeholders and worked with ERCOT staff to understand the relevant aspects of their operations and market data. We performed probabilistic simulation analyses of prices, investment costs, and reliability. Our findings informed a Public Utility Commission of Texas (PUCT) proceeding in which Brattle experts filed comments and presented at several workshops. The findings ultimately led to ERCOT’s development of the Operating Reserve Demand Curve (ORDC).

ERCOT economically-optimal reserve margin

For the PUCT and ERCOT, Brattle co-authored a report estimating the economically-optimal reserve margin compared to various reliability-based reserve margins, and evaluated the cost and uncertainty of an energy-only and a potential capacity market in ERCOT. The simulations incorporated a detailed representation of the Texas power market, including intermittent wind and solar generation, operating reserves, different types of demand response, the full range of emergency procedures (such as operating reserve deletion), scarcity pricing provisions, and load-shed events.

FERC economics of reliability study

Brattle experts conducted a study on the economics of reliability for the Federal Energy Regulatory Commission (FERC) in conjunction with Astrape Consulting. One component of the study was the development and demonstration of an approach for estimating a value-based demand curve for capacity. The study also addressed other aspects of the economics of reliability, including: (a) the sensitivity of reliability-based reserve margins to study assumptions, reliability metric definitions, and system conditions; (b) the “economically optimal” reserve margin compared to reliability-based reserve margins; (c) the impacts of varying demand response (DR) penetration, wind penetration, load forecast error, and intertie levels; and (d) the price, system cost, and customer cost impacts of different energy and capacity market designs.

News & Knowledge