Constrained multi-objective optimization of process design parameters in settings with scarce data: an application to adhesive bonding

Alejandro Morales-Hernández,Sebastian Rojas Gonzalez,Inneke Van Nieuwenhuyse,Ivo Couckuyt,Jeroen Jordens,Maarten Witters,Bart Van Doninck

DOI: https://doi.org/10.48550/arXiv.2112.08760

2023-04-10

Abstract:Adhesive joints are increasingly used in industry for a wide variety of applications because of their favorable characteristics such as high strength-to-weight ratio, design flexibility, limited stress concentrations, planar force transfer, good damage tolerance, and fatigue resistance. Finding the optimal process parameters for an adhesive bonding process is challenging: the optimization is inherently multi-objective (aiming to maximize break strength while minimizing cost), constrained (the process should not result in any visual damage to the materials, and stress tests should not result in failures that are adhesion-related), and uncertain (testing the same process parameters several times may lead to different break strengths). Real-life physical experiments in the lab are expensive to perform. Traditional evolutionary approaches (such as genetic algorithms) are then ill-suited to solve the problem, due to the prohibitive amount of experiments required for evaluation. Although Bayesian optimization-based algorithms are preferred to solve such expensive problems, few methods consider the optimization of more than one (noisy) objective and several constraints at the same time. In this research, we successfully applied specific machine learning techniques (Gaussian Process Regression) to emulate the objective and constraint functions based on a limited amount of experimental data. The techniques are embedded in a Bayesian optimization algorithm, which succeeds in detecting Pareto-optimal process settings in a highly efficient way (i.e., requiring a limited number of physical experiments).

Neural and Evolutionary Computing,Machine Learning

What problem does this paper attempt to address?

### What problems does this paper attempt to solve?

This paper aims to solve the problem of multi - objective optimization in the bonding process, especially in the case of scarce data. Specifically, the paper focuses on how to find the optimal bonding process parameters in practical applications to meet multiple conflicting objectives and constraints.

#### Specific problem description:

1. **Multi - objective optimization**:
- One of the objectives is to maximize the tensile strength (TS).
- Another objective is to minimize the production cost (PC).
These two objectives are usually in conflict, that is, increasing the bonding strength may increase the cost, and vice versa.

2. **Constraints**:
- The bonding process should not cause any visible damage to the material.
- The stress test should not cause bonding - related failures.

3. **Uncertainty**:
- Testing the same set of process parameters multiple times may result in different fracture strengths, indicating that the experimental results are noisy.

4. **High experimental cost**:
- Physical experiments in the laboratory are very expensive, so an efficient method is required to reduce the number of required experiments.

#### Solution:

To address the above challenges, the paper proposes a method based on Bayesian optimization (BO), combined with Gaussian process regression (GPR) to model the objective function and the constraint function. This method can efficiently find Pareto - optimal solutions with a limited amount of data, thereby reducing the dependence on a large number of physical experiments.

#### Main contributions:

1. **Improved acquisition function**:
- A proposed acquisition function that combines expected improvement (EI) and constraint feasibility, which is suitable for black - box, noisy, and computationally expensive objective functions and constraints.

2. **Handling noise**:
- The Gaussian process model (GP) is used to explicitly consider the noise in the experimental results, especially the input - dependent heterogeneous noise.

3. **Efficient multi - objective optimization algorithm**:
- Two new Bayesian multi - objective optimization algorithms, cMEI - SK and cEHVI - SK, are developed, using different acquisition functions respectively, which can find approximate Pareto - optimal solutions within a very limited number of experiments.

Through these methods, the paper shows how to achieve efficient and low - cost multi - objective optimization in the bonding process, providing a valuable reference for similar problems.