Traditional probabilistic risk assessment (PRA), as formalized in the reactor safety study and subsequent industry practice,
has provided a foundational framework for quantifying nuclear power plant risk. However, its mathematical structure relies on assumptions that are increasingly misaligned with modern regulatory and operational realities, including infinite time horizons, steady-state availability metrics, memoryless (e.g., Poisson distribution) initiating event arrivals, and static fault tree representations of system behavior.

This work synthesizes recent developments that recast risk quantification as a time-domain stochastic process problem using marked point processes (MPP) and discrete-event simulation (DES). In this formulation, system behavior is represented as a stochastic state process evolving through transitions driven by endogenous failures, exogenous hazards, maintenance, testing, and operational decisions, with risk emerging from the temporal interaction of these processes rather than from static combinatorial logic.

The framework relaxes key PRA assumptions by introducing finite-horizon evaluation, state-dependent initiating event processes, explicit modeling of repairable systems with imperfect maintenance, and dependencies across components, failure modes, and external stressors. Maintenance is represented as a stochastic intervention with efficacy, enabling modeling of both degraded and “better-than-new” outcomes and capturing the effect of corrective action on future failure exposure. This formulation supports explicit treatment of maintenance subduing, common-cause influences, environmental forcing, and configuration-specific vulnerabilities within a unified analytic structure.

A key distinction from classical PRA is that the proposed framework does not rely on rare-event approximations or minimal
cut set representations as the primary basis for risk quantification. Instead, accident occurrence is modeled as a counting process derived from underlying system state trajectories, allowing direct evaluation of time-to-accident distributions, exposure durations, and conditional risk under evolving configurations. This enables consistent treatment of overlapping exposures, time-varying configurations, dependencies, and nonstationary operating conditions that are difficult to represent within traditional steady-state formulations.

These advances introduce challenges, including increased computational demand for high-fidelity rare-event simulation and
expanded data requirements for time-dependent failure behavior, repair duration distributions, maintenance effectiveness, and dependency modeling. They also raise questions regarding appropriate model abstraction, interpretability, and consistency with regulatory decision processes. More fundamentally, by making time-dependent uncertainty and dependency structures explicit, the framework may shift emphasis in regulatory evaluation from strict prevention of core damage toward robust mitigation and containment of its consequences. Historical experience, including the ECCS hearings, reflects recognition that uncertainty in protection system performance can motivate increased reliance on containment as a primary barrier to public risk. Within an MPP/DES formulation, such tradeoffs can be evaluated directly by quantifying how design and operational choices influence both accident occurrence and post-accident consequence exposure over time. This perspective is consistent with a regulatory strategy in which established industry standards are relied upon for core damage prevention, while containment systems are explicitly designed and evaluated under the assumption that core damage has occurred, thereby ensuring robustness against bounding accident conditions even if their likelihood remains low.

The transition to MPP/DES has important implications for regulation. The U.S. Nuclear Regulatory Commission (NRC),
acting as the final arbiter of acceptable safety, must evaluate whether design and operational choices satisfy statutory standards for adequate protection. The hand rule provides a useful lens for this evaluation, but classical PRA metrics based on long-run frequency can obscure time-dependent and configuration-specific risk. In contrast, MPP/DES supports direct estimation of quantities more closely aligned with expected loss and with the observable performance of engineered protections over time.

Recent judicial developments, including Loper Bright Enterprises v. Raimondo, further emphasize the need for transparent and technically grounded regulatory analysis. By reducing judicial deference to agency interpretations, such developments increase the importance of analytical methods that clearly connect assumptions, system behavior, and risk outcomes. Taken together, these considerations suggest that MPP/DES-based risk quantification represents not only a methodological advancement, but a necessary step toward aligning engineering analysis with the legal and institutional structure of nuclear safety regulation.