Emulation of utility functions over a set of permutations: sequencing reliability growth tasks

TL;DR

This paper introduces a surrogate modeling approach to efficiently emulate the expected utility over permutations in Bayesian experimental design, enabling better sequencing of tasks under computational constraints.

Contribution

It proposes a parametric probabilistic model to approximate expected utility over permutations, improving efficiency in large-scale sequencing problems.

Findings

The surrogate model accurately predicts expected utility with fewer evaluations.

The approach effectively sequences reliability growth tasks in hardware development.

Simulation results demonstrate the method's robustness and practical utility.

Abstract

We consider Bayesian design of experiments problems in which we maximise the prior expectation of a utility function over a set of permutations, for example when sequencing a number of tasks to perform. When the number of tasks is large and the expected utility is expensive to compute, it may be unreasonable or infeasible to evaluate the expected utility of all permutations. We propose an approach to emulate the expected utility using a surrogate function based on a parametric probabilistic model for permutations. The surrogate function is fitted by maximising the correlation with the expected utility over a set of training points. We propose a suitable transformation of the expected utility to improve the fit. We provide results linking the correlation between the two functions and the number of expected utility evaluations to undertake. The approach is applied to the sequencing of reliability growth tasks in the development of hardware systems, in which there is a large number of potential tasks to perform and engineers are interested in meeting a reliability target subject to minimising costs and time. An illustrative example shows how the approach can be used and a simulation study demonstrates the performance of the approach more generally.