This paper presents enhancements to a surrogateassisted evolutionary optimization framework proposed earlier in the literature for solving computationally expensive design problems on a limited computational budget [1]. The main idea of our former framework was to couple evolutionary algorithms with a feasible sequential quadratic programming solver in the spirit of Lamarckian learning, including a trustregion approach for interleaving the true fitness function with computationally cheap local surrogate models during gradientbased search. In this paper, we propose a hierarchical surrogateassisted evolutionary optimization framework for accelerating the convergence rate of the original surrogate-assisted evolutionary optimization framework. Instead of using the exact high-fidelity fitness function during evolutionary search, a Kriging global surrogate model is used to screen the population for individuals that will undergo Lamarckian learning. Numerical results are presented for two multi-modal benchmark test functions to show that the proposed approach leads to a further acceleration of the evolutionary search process.

Abstract: Many real-world manufacturing problems are too complex to be modelled analytically. In these scenarios simulation-based optimisation is a powerful tool to determine optimal system settings. While traditional optimisation methods have been unable to cope with the complexities of many real-world problems approached by simulation, soft computing techniques have proven to be highly useful. This paper describes how simulation and soft computing techniques have been combined and successfully used to optimise a manufacturing system at Volvo Aero in Trollhättan.

Abstract The paper addresses key problems pertaining to the commonly used evolutionary approach to the search for optimal catalysts in chemical engineering. These are on the one hand the insufficient dealing in existing implementations of genetic algorithms with mixed optimization, which plays a crucial role in catalysis, on the other hand the narrow scope of genetic algorithms developed specifically for searching optimal catalyst. The paper proposes an approach to constrained mixed optimization based on formulating a separate linearly-constrained continuous optimization task for each combination of values of the discrete variables. Then, discrete optimization on the set of nonempty polyhedra describing the feasible solutions of those tasks is performed, followed by solving those tasks for each individual of the resulting population of polyhedra. To avoid computationally expensive checking of the non-emptiness of individual polyhedra, the set of polyhedra is first partitioned into equivalence classes such that only one representative from each class needs to be checked. Finally, the paper outlines a program generator automatically generating problem-tailored genetic algorithms from descriptions of optimization tasks in a specific description language, which employs the proposed approach to constrained mixed optimization.

Genetic algorithms (GAs) used in complex optimization domains usually need to perform a large number of fitness function evaluations in order to get near-optimal solutions. In real world application domains such as the engineering design problems, such evaluations might be extremely expensive computationally. It is therefore common to estimate or approximate the fitness using certain methods. A popular method is to construct a so called surrogate or meta-model to approximate the original fitness function, which can simulate the behavior of the original fitness function but can be evaluated much faster. It is usually difficult to determine which approximate model should be used and/or what the frequency of usage should be. The answer also varies depending on the individual problem. To solve this problem, an adaptive fitness approximation GA (ASAGA) is presented. ASAGA adaptively chooses the appropriate model type; adaptively adjusts the model complexity and the frequency of model usage according to time spent and model accuracy. ASAGA also introduces a stochastic penalty function method to handle constraints. Experiments show that ASAGA outperforms non-adaptive surrogate-assisted GAs with statistical significance.

serious game for understanding artificial intelligence in production optimization

Abstract—In this work, we propose a new genetic programming algorithm with local search strategies, named Memetic Genetic Programming(MGP), for classification problems. MGP aims to acquire a classifier with large Area Under the ROC Curve (AUC), which has been proved to be a better performance metric for traditionally used metrics (e.g., classification accuracy). Three new points are presented in our new algorithm. First, a new representation called statistical genetic decision tree (SGDT) for GP is proposed on the basis of Genetic Decision Tree (GDT). Second, a new fitness function is designed by using statistic information from SGDT. Third, the concept of memetic computing is introduced into SGDT. As a result, the MGP is equipped with a local search method based on the training algorithms for decision trees. The efficacy of the MGP is empirically justified against a number of relevant approaches. Index Terms—Genetic Programming; Memetic Algorithm; AUC; Classification

We will first attempt to position multilevel model optimization within the more general framework of MDO. As detailed in Chapters 8 (theory) and 14 (algorithmic aspects), in the general MDO approach, optimization algorithms and simulation models appear to be decoupled: we will proceed with this assumption, while being aware