Loss of Load Probability (LOLP)

LOLP is formally defined as the probability that the available generation capacity is less than the system load over a specified time interval. It is a dimensionless value between 0 and 1, often expressed as a percentage or in days per year.

• Mathematical form: LOLP = P(Available Capacity < Load)
• Time horizon: Typically calculated for peak load hours, a single day, or an entire year
• Not a duration metric: LOLP measures the likelihood of a shortfall, not its magnitude or how long it lasts

LOLP is the primary metric in generation adequacy studies, which evaluate whether a power system has sufficient installed capacity to meet demand. It accounts for both forced outage rates (FOR) of individual generators and load forecast uncertainty.

• Each generator is modeled as a two-state Markov process: available or on forced outage
• The capacity outage probability table (COPT) aggregates all unit states into a discrete probability distribution
• LOLP is the cumulative probability of all outage states exceeding the reserve margin

The classic LOLP calculation uses recursive convolution to build the capacity outage probability table. Each generating unit's outage distribution is convolved with the existing table, producing a new cumulative distribution.

• Unit addition algorithm: P_new(X) = (1 - FOR) × P_existing(X - C) + FOR × P_existing(X)
• Where C is unit capacity and FOR is the forced outage rate
• Modern implementations use Monte Carlo simulation or frequency and duration methods for complex systems with energy-limited resources

LOLP is closely related to Loss of Load Expectation (LOLE), which converts probability into a tangible time metric. LOLE is the expected number of hours or days per year that load will exceed available capacity.

• Conversion: LOLE (days/year) = LOLP × 365 (if LOLP is a daily peak probability)
• Industry standard: The North American Electric Reliability Corporation (NERC) uses a 1-day-in-10-years LOLE criterion (0.1 days/year)
• LOLP provides the instantaneous risk; LOLE aggregates it into a planning target

Traditional LOLP has known limitations when applied to systems with high variable renewable energy (VRE) penetration. It does not capture the temporal correlation of wind and solar output or the flexibility constraints of thermal generators.

• No ramp rate modeling: LOLP treats each hour independently, ignoring inter-hour ramping capability
• No energy storage: Batteries and pumped hydro are energy-limited resources that do not fit the two-state availability model
• Extensions required: Modern adequacy studies use sequential Monte Carlo or stochastic unit commitment to capture chronological constraints

LOLP serves as a regulatory threshold for resource adequacy in many jurisdictions. Utilities and system operators must demonstrate that their resource plans maintain LOLP below a mandated ceiling.

• Resource planning: LOLP drives the effective load carrying capability (ELCC) calculation for new generators
• Capacity markets: LOLP informs the demand curve in forward capacity auctions (e.g., PJM, ISO New England)
• Probabilistic reserve margins: LOLP replaces deterministic reserve margin rules with risk-based targets

The expected number of days or hours per year when the available generation capacity is insufficient to meet the daily peak load. While LOLP measures the probability of failure at any given moment, LOLE aggregates this probability over a defined period—typically a year—to produce a tangible duration metric. A common planning standard is a LOLE of 0.1 days per year, or one day of expected shortfall every ten years. LOLE is calculated by summing the daily LOLP values over all days in the study horizon, converting a dimensionless probability into an actionable operational metric for capacity planners.

A severity metric that quantifies the total megawatt-hours of demand that cannot be supplied due to generation inadequacy over a given period. Unlike LOLP, which only measures the likelihood of a shortfall, EUE captures the magnitude of the deficit. It is computed by integrating the product of the probability of each capacity outage state and the corresponding load curtailment. EUE is critical for economic reliability planning because it enables cost-benefit analysis: the cost of adding new generation capacity is weighed against the economic value of the unserved energy it prevents.

The additional load that a new generator—particularly a variable renewable resource like solar or wind—can serve without degrading the system's reliability target. ELCC is determined by iteratively adding the generator to a reliability model and adjusting the load until the LOLP returns to the baseline value. A wind farm with a nameplate capacity of 100 MW might have an ELCC of only 15 MW, reflecting its intermittent contribution to resource adequacy. This metric directly links probabilistic reliability analysis to capacity market accreditation.

A fundamental input to LOLP calculation that enumerates all possible combinations of generator forced outages and their associated probabilities. The COPT is constructed using the binomial distribution for identical units or recursive convolution for a heterogeneous fleet. Each row represents a discrete capacity outage state, with columns for the amount of capacity unavailable and the exact probability of that state occurring. The LOLP is computed by summing the probabilities of all COPT states where the remaining available capacity falls below the system load.

The probability that a generating unit will be unavailable due to an unplanned failure at any given time. FOR is the primary stochastic input driving the COPT and, consequently, LOLP. It is calculated as the ratio of forced outage hours to the sum of forced outage hours and service hours. A typical coal plant might have a FOR of 6-8%, while a combined-cycle gas turbine may be as low as 2-3%. Accurate FOR estimation from historical operational data is essential for credible reliability modeling.

The estimated economic cost to consumers for each megawatt-hour of electricity that cannot be supplied. VOLL translates the physical reliability metric LOLP into monetary terms, enabling optimal investment decisions. It varies by sector—residential VOLL is typically much higher than industrial—and by outage duration. When combined with EUE, VOLL yields the total economic damage of supply interruptions, which is then compared against the cost of new capacity to determine the economically efficient reliability target.