Non-determinative Risk Simulation via Neural Network
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current risk assessment models in enterprise risk management are limited by their determinative formulas, which do not allow for new information to be gained through simulation, making it difficult to effectively mitigate risks and understand the impact of changes on risk scores.
Innovation Solution
A compliance management system that utilizes a non-linear statistical data model, such as a neural network, to analyze multiple sources of information and calculate risk scores, allowing for risk simulation and mitigation scenarios to be studied, providing insights not available with traditional determinative models.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If determinative risk formulas and models are used to calculate risks, then the calculation process is simple and straightforward, but no new information can be gained by attempting to mitigate risks through simulation
Solution Approach 1:
The patent transforms the fixed, determinative risk model parameters into variable, probabilistic parameters that can be simulated. By changing from deterministic values to probability distributions, the system enables simulation while maintaining practical usability through standardized risk factors and control effectiveness parameters.
Solution Approach 2:
The patent creates a virtual copy of the risk assessment system that can be simulated without affecting the actual system. This virtual model replicates the risk calculation logic and allows repeated simulations with different inputs to gain information about risk mitigation strategies.
2Ease of operation
If traditional determinative models are used, then the model structure is simple and easy to understand, but the ability to provide actionable insights for risk mitigation is limited
Solution Approach 1:
The patent implements feedback loops where simulation results feed back into the risk assessment process. The system compares simulated outcomes with current risk states, providing actionable feedback on which controls to adjust and how changes will impact overall risk, thereby enabling continuous improvement of risk mitigation strategies.
Solution Approach 2:
The patent transforms static risk models into dynamic systems that can adapt to different scenarios. By allowing parameters to vary through simulation and providing real-time feedback on risk changes, the system becomes dynamic while maintaining clarity through structured presentation of results and their implications for mitigation.
Data Source
AI summary
Simulating risk circumstances can reveal new information to risk assessment personnel about how to mitigate risk. In one embodiment, the present invention includes selecting an asset from a plurality of heterogeneous assets of a business enterprise. The user can then input a plurality of simulated risk factors for the selected asset into the system which receives this input. The risk assessment system can then generate a non-determinative simulated risk score using the simulated risk factors, the simulated risk score being a simulated measure of risk associated with the selected asset if the selected asset were to be associated with the plurality of simulated risk factors.