Time-Variant Reliability Dataset for the Nipigon Slope under Rainfall Events

This dataset is organized into two sheets: input and output, and is intended for time-variant reliability analysis of an unsaturated slope subjected to rainfall. The data was generated through numerical simulations using GeoStudio software to evaluate slope stability over time under a wide range of conditions.

The input sheet contains 1000 simulation trials, each defined by a unique set of 26 randomly generated variables. These variables represent the material and hydraulic properties of a three-layered soil slope, including cohesion, friction angle, unit weight, saturated and unsaturated hydraulic conductivity, and water retention characteristics for each layer. Additionally, rainfall duration and rainfall intensity are included among the inputs to simulate different rainfall events. The random values were drawn from realistic ranges based on typical soil behavior and field data.

The output sheet contains the corresponding Factor of Safety (FOS) values over time for each trial. For every simulation, the FOS was computed at multiple time steps, capturing the evolution of slope stability as rainfall infiltrated the unsaturated soil profile. Each row corresponds to one trial from the input sheet, and each column represents the FOS at a specific time step during the simulated event.

The modeled slope is based on the Nipigon slope, located in northern Ontario, Canada, along the Trans-Canada Highway. This site is known for its fine-grained, moisture-sensitive soils and a history of rainfall-induced instability, making it a representative case for studying unsaturated slope behavior under variable environmental loading.

This dataset is a valuable resource for researchers and engineers in geotechnical and reliability fields. It supports the development and validation of meta-models, such as artificial neural networks, and enables Monte Carlo simulation, sensitivity analysis, and probabilistic slope failure assessment. By incorporating random variation across a wide range of inputs and capturing time-dependent FOS responses, this dataset enhances understanding of the complex interactions between soil behavior, rainfall, and slope stability—ultimately contributing to better prediction and mitigation of slope failure risks.