Uncertainty Estimation of the Optimal Decision with Application to Cure Process Optimization

2025-07-29

AI Summary

In simulation-driven manufacturing, process optimization based on limited simulation data suffers from uncertainty in optimal decisions due to restrictive modeling assumptions and data scarcity; moreover, the sensitivity of the objective function—and thus of the optimal solution—to input parameters remains unquantified systematically. Method: We propose a surrogate modeling framework integrating uncertainty propagation with local sensitivity analysis: a Gaussian process (GP) models the objective function; posterior sampling generates a distribution of optimal solutions; and a gradient-based sensitivity metric quantifies how perturbations in key parameters affect the location of the optimal solution. Contribution/Results: Validated on composite material curing process optimization, our method significantly enhances the reliability and robustness of data-driven decision-making. It provides a generalizable, uncertainty-aware optimization methodology for simulation-intensive manufacturing, enabling principled quantification of both solution uncertainty and parameter sensitivity under data constraints.

Technology Category

• Search and Optimization: Sampling/Simulation-based Search
• Reasoning under Uncertainty: Stochastic Optimization
• Machine Learning: Optimization

Application Category

• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems
• Web Mining and Content Analysis: Web data generation and simulation
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Decision-making in manufacturing often involves optimizing key process parameters using data collected from simulation experiments. Gaussian processes are widely used to surrogate the underlying system and guide optimization. Uncertainty often inherent in the decisions given by the surrogate model due to limited data and model assumptions. This paper proposes a surrogate model-based framework for estimating the uncertainty of optimal decisions and analyzing its sensitivity with respect to the objective function. The proposed approach is applied to the composite cure process simulation in manufacturing.

Problem

• Estimating uncertainty in optimal manufacturing decisions
• Analyzing sensitivity of decisions to objective functions
• Optimizing composite cure process using surrogate models

Innovation

• Gaussian processes surrogate model for optimization
• Estimates uncertainty of optimal decisions
• Analyzes sensitivity of objective function