Search

The accurate throughput estimation of a production line steadily attracts research interest due to its high impact on their economic operation affecting the world production of goods. Since Artificial Intelligence (AI) is embedded in production processes leading to Industry 4.0, it is as well used for throughput estimation. Neural Networks (NN) are a popular and well-established and tested intelligent methodology capable of performing at an excellent level as function approximation techniques for complex domains. In this paper different neural network architectures as well as deep learning approaches have been tested on datasets corresponding to accurate solutions of throughput taken from exact algorithms existing in literature for relatively small size serial production lines without buffers. Three main subproblems are studied in the paper, those of three, four and five workstations without intermediate buffers. The idea is to train a neural network architecture over each dataset corresponding to a specific problem and obtain generalized models of high accuracy, able to accurately calculate throughput for unknown settings and parameters of similar characteristics and parameters within the training range (i.e., the mean service rate) but also out of it. Performance deteriorates as we move away from the training range, a fact that leads to the conclusion that the expert should know the basic operating characteristics and properties of a specific production line, prior to setting the training interval into which the neural network should be trained, and then be used for throughput approximation. It is interesting that the datasets used for training are very small compared to those used for validation, which proves the high neural network generalization power. The performance of the neural network approach is compared to a previously applied genetic programming (GP) approach and it proves that it performs competitively, and in some cases slightly better within the trai

This paper focuses on approaching reliability–redundancy allocation problems, using hybrid schemes that consist of individual nature inspired algorithms. The aim is to investigate if hybridization is an efficient way to approach problems with multiple goals. Therefore, known algorithms that have been successfully applied to reliability and redundancy allocation problems in literature are implemented as part of hybrid schemes in this study. The idea behind that is to study whether an efficient hybrid scheme is the outcome of the hybridization of (individual) efficient algorithms. The performance of the nine proposed schemes and the individual algorithms is tested on ten well-known artificial and real-world case studies, from the field of reliability engineering. The numerical results are compared to others from literature highlighting the efficiency of the proposed hybrid schemes and consequently, support the hypothesis that hybridization can enhance the performance of optimization methods. The experimental results demonstrate that among the nine hybrid schemes, the one consisting of Bat Algorithm and Firefly Algorithm shows the best performance.

One of the most well-known problems in combinatorial optimization is the Vehicle Routing Problem (VRP). Significant research has been done around this problem in two different perspectives: investigating new solving approaches, and studying variants of VRP which take into consideration multiple restrictions and constraints. One of such versions is the Electric Vehicle Routing Problem (EVRP), whose main objective is to find the optimal route of a fleet of electric vehicles, taking into account the locations of charging stations and the battery consumption of the mobile units. The aim of this study is threefold: (a) to perform a brief literature review on meta-heuristic approaches applied to the EVRP, (b) to offer insights on the available data instances for this problem, and (c) to discuss on the results of an experimental benchmark aimed at comparing different meta-heuristic approaches over diverse EVRP instances, including the proposal and evaluation of a novel Ant Colony Optimization approach.

The constraints and the vast solution space of operational research optimization problems make them hard to cope with. However, Computational Intelligence, and especially Nature-Inspired Algorithms, has been a useful tool to tackle hard and large space optimization problems. In this paper, a very consistent and effective hybrid optimization scheme to tackle cardinality constrained portfolio optimization problems is presented. This scheme consists of two nature-inspired algorithms, i.e. Sonar Inspired Optimization algorithm and Genetic Algorithm. Also, the incorporation of heuristic information, i.e. an expert’s knowledge, etc., to the overall performance of the hybrid scheme is tested and compared to previous studies. More specifically, under the framework of a financial portfolio optimization problem, the heuristic information-enhanced hybrid scheme manages to reach a new optimal solution. Additionally, a comparison of the proposed hybrid scheme with other hybrid schemes applied to the same problem with the same data is performed.

In the last decade, we observe an increasing number of nature-inspired optimization algorithms, with authors often claiming their novelty and their capabilities of acting as powerful optimization techniques. However, a considerable number of these algorithms do not seem to draw inspiration from nature or to incorporate successful tactics, laws, or practices existing in natural systems, while also some of them have never been applied in any optimization field, since their first appearance in literature. This paper presents some interesting findings that have emerged after the extensive study of most of the existing nature-inspired algorithms. The need for irrationally introducing new nature inspired intelligent (NII) algorithms in literature is also questioned and possible drawbacks of NII algorithms met in literature are discussed. In addition, guidelines for the development of new nature-inspired algorithms are proposed, in an attempt to limit the misleading appearance of variation of metaheuristics as nature inspired optimization algorithms.

In the recent years, extensive discussion takes place in literature, on the effectiveness of meta-heuristics, and especially Nature Inspired Algorithms. Usually, authors state that such an approach should embody a well-balanced exploration and exploitation strategy. Sonar Inspired Optimization (SIO) is a recently presented algorithm, which counts already a number of successful real-world applications. Its novel mechanisms provide this equilibrium between exploration and exploitation, as it has been stated in previous studies. In this work, authors prove that this equilibrium exists and also, it is one of the main reasons behind the high quality performance of SIO.

Optimal resource allocation is a challenging problem which is faced by Project Managers. In large projects, where multiple resources have to be properly allocated taking into consideration the total cost of project and the time needed to be completed, the optimization problem becomes even more difficult. Therefore, intelligent techniques have been widely used to cope with such demanding problems. This study applies a nature-inspired intelligent algorithm, i.e. Sonar Inspired Optimization (SIO), to face the Resource Leveling problem of a real world project, i.e. a High Voltage Alternating Current Submarine Link Terminal Station. The specific application domain is a NP-hard optimization problem as it can receive a very large number of possible solutions. Furthermore, a hybrid scheme of this algorithm with Simulated Annealing (SIO-SA) is used to improve the performance of SIO. Comparative results show that both approaches (SIO and SIO-SA) perform almost equally well.

For many decades, Machine Learning made it possible for humans to map the patterns that govern interpolating problems and also, provided methods to cluster and classify big amount of uncharted data. In recent years, optimization problems which can be mathematically formulated and are hard to be solved with simple or naïve heuristic methods brought up the need for new methods, namely Evolutionary Strategies. These methods are inspired by strategies that are met in flora and fauna in nature. However, a lot of these methods are called nature-inspired when there is no such inspiration in their algorithmic model. Furthermore, even more evolutionary schemes are presented each year, but the lack of applications makes them of no actual utility. In this chapter, all Swarm Intelligence methods as far as the methods that are not inspired by swarms, flocks or groups, but still derive their inspiration by animal behaviors are collected. The applications of these two sub-categories are investigated and some preliminary findings are presented to highlight some main points for Nature Inspired Intelligence utility.

One of the upcoming categories of Computational Intelligence (CI) is meta-heuristic schemes, which derive their intelligence from strategies that are met in nature, namely Nature Inspired Algorithms. These algorithms are used in various optimization problems because of their ability to cope with multi-objective problems and solve difficult constraint optimization problems. In this work, the performance of Sonar Inspired Optimization (SIO) is tested in a non-smooth, non-convex multi-objective Energy problem, namely the Economic Emissions Load Dispatch (EELD) problem. The research hypothesis was that this new nature-inspired method would provide better solutions because of its mechanisms. The algorithm manages to deal with constraints, namely Valve-point Effect and Multi-fuel Operation, and produces only feasible solutions, which satisfy power demand and operating limits of the system examined. Also, with a lot less number of agents manages to be very competitive against other meta-heuristics, such as hybrid schemes and established nature inspired algorithms. Furthermore, the proposed scheme outperforms several methods derived from literature.

These data contain a comprehensive collection of all Nature-Inspired Algorithms. This collection is a result of two corresponding surveys, where all Nature-Inspired Algorithms that have been published to-date were gathered and preliminary data acquired. The rapidly increasing number of nature-inspired approaches makes it hard for interested researchers to keep up. Moreover, a proper taxonomy is necessary, based on specific features of the algorithms. Different taxonomies and useful insight into the application areas that the algorithms have coped with is given through these data. This article provides a detailed description of the above mentioned collection.

Recently, a new Nature Inspired Intelligent scheme has been proposed and presented, named Sonar Inspired Optimization (SIO). This algorithm is inspired by the SONAR mechanism, which is used by Warships to detect targets and avoid mines. In this paper, improvements have been done to the SIO approach in an attempt to increase the performance of the algorithm. Also, results from experiments in known constrained Engineering applications are presented and discussed. SIO tackles with these problems, managing to overcome the performance of other Nature Inspired metaheuristics, heuristics and mathematical approaches in most of the cases.

One of the problems that Machine Learning (ML) algorithms face in classification tasks is the Curse of Dimensionality, which refers to the sensitivity of their performance to the data dimensionality. The solution to this problem is to select the features that are of higher importance for the model produced. To improve the performance of a classifier, various meta-heuristic algorithms have been implemented due to their ability to provide optimal solutions in problems that there are multiple candidate solutions, such as in Feature Selection (FS). In this study, Sonar Inspired Optimization (SIO) algorithm is used to perform FS in order to improve the performance of a state-of-the-art classifier, i.e. k-NN. SIO's performance is compared with other nature-inspired meta-heuristic algorithms that have been used for the same task.

Resource Leveling is a constrained problem with a large solution space, especially when the projects have numerous tasks. This makes the problem hard to tackle. In this study, a novel algorithm named sonar inspired optimization (SIO) belonging in Nature Inspired Intelligence is applied in both benchmark and artificial resource-leveling problems to investigate its performance. Results are compared to those derived from a Hybrid Genetic Algorithm (HGA) previously developed. Experimental findings show that the proposed algorithm is very promising as in most cases the obtained solutions prove superior or equally good to those of HGA and other competitive approaches. An additional approach of the proposed metaheuristic is that it generates only feasible solutions. The algorithm has been implemented in different programming languages so that in future applications will be user friendly for project managers.

Estimating the performance of a production line is a difficult problem because of the enormous number of states that exist when analyzing such systems. In addition to the methods developed to address the problem, it is very useful to have a formula linking the characteristics of the line to its performance. Three cases of sort serial production lines with parallel and identical machines in each workstation are examined in this paper. By using a combinational method that applies genetic programming (GP) and an innovative nature inspired method, named sonar inspired optimization (SIO) to improve the results, three models are derived to obtain the throughput of the corresponding lines. Further work will take place because results derived in this paper are encouraging.

In previous studies, nature-inspired algorithms have been implemented in order to tackle hard NP-optimization problems, in the financial domain. Specifically, the task of finding optimal combination of assets with the aim of efficiently allocating your available capital is of major concern. One of the main reasons, which justifies the difficulties entailed in this problem, is the high level of uncertainty in the financial markets and not only. As mentioned above, artificial intelligent algorithms may provide a solution to this task. However, there is one major drawback concerning these techniques: the large number of open parameters. The aim of this study is twofold. Firstly, results from extended simulations are presented regarding the application of a specific hybrid nature-inspired metaheuristic in a particular formulation of the financial portfolio optimization problem. The main focus is on presenting comparative results regarding the performance of the proposed scheme for various configuration settings. Secondly, it is our intend to enhance the hybrid scheme's performance by incorporating intelligent searching components such as other metaheuristics (simulated annealing).