Data clustering is commonly employed in many disciplines. The aim of to partition a set data into clusters, which objects within the same cluster are similar and dissimilar other that belong different clusters. Over past decade, evolutionary algorithm has been used solve problems. This study presents novel based on simplified swarm optimization, an emerging population-based stochastic optimization approach with advantages simplicity, efficiency, flexibility. combines variable vibrating...

Nowadays, cloud computing and big data are changing the enterprise. Cloud computing, as a new business mode, distributes tasks across resource pools made up of large number computers for large-scale calculation. In current research on task assignment problem most scholars consider single-objective programming, example minimizing cost or makespan. However, many other factors can influence quality service. Therefore, in order to adapt development practical applications, multi-objective...

Abstract With the dramatic growth of data volume, cloud computing structure has faced a severe difficulty, latency. To deal with this problem, researchers have proposed fog structure, which can successfully release computation loads from one datacenter to multiple local devices. Hence, tasks will be processed at device and avoid transmitting is not cost-effective, results transmitted users immediately. The main differences task scheduling between are processors specifications, such as...

For realizing smart manufacturing, recent studies have been focusing on the capabilities of digital twins (DTs) that offer a virtual portrayal process or system and thereby facilitate real-time monitoring, analysis, optimization. DTs construct “digital model” factories by using advanced technologies such as sensors, Internet Things (IoT), artificial intelligence (AI) techniques, optimization processes for executing critical decisions enhancing productivity predicting future events....

Convolutional neural networks (CNNs) are widely used in image recognition. Numerous CNN models, such as LeNet, AlexNet, VGG, ResNet, and GoogLeNet, have been proposed by increasing the number of layers, to improve performance CNNs. However, deteriorates beyond a certain layers. Hence, hyperparameter optimisation is more efficient way To validate this concept, new algorithm based on simplified swarm optimise hyperparameters simplest model, which LeNet. The results experiments conducted MNIST,...

In the 21th century, importance of integration among enterprises has been raised. Many and managements start to be conscious supply chain management. management, reducing operating cost satisfying customer demand are most important things. However, products may spoilt during delivery due collisions, traffic accident, weather factor, theft so on. Hence, in this paper, we consider deterioration effect a three-stage deteriorated network with mathematical model. A novel artificial intelligence...

Abstract Fog computing is an emerging technology that can reduce the load on cloud system and decentralize resource, thus increasing response speed reducing time delay. More environments in order to achieve intelligence, collect massive amounts of data though IOT devices. Considering deploying fog these make faster more robust. This work creates a simulated factory consisting center, gateways, devices, edge different types sensors. We build integer programming model apply metaheuristic...

Single Row Facility Layout Problem (SRFLP) is a NP-Complete permutation problem which has been widely studied in the academic field order to improve efficiency on production site. The main purpose of SRFLP find an optimal given number rectangular facilities with minimum total cost by arranging them along straight line. In this paper, we use Simplified Swarm Optimization algorithm effective local search near-optimal solution for SRFLP. comparison computational results benchmark problems...

In recent years, Convolutional Neural Networks (CNNs) have been widely used in image recognition due to their aptitude large scale processing. The CNN uses Back-propagation (BP) train weights and biases, which turn makes the error consistently smaller. most common optimizers that a BP algorithm are Stochastic Gradient Decent (SGD), Adam, Adadelta. These optimizers, however, proved fall easily into regional optimal solution. Little research has conducted on application of Soft Computing fix...

In recent decade, reliability has been an important factor which may affect the performance of system. order to enhance system reliability, redundancy allocation problem (RAP) is becoming increasingly tool in stages planning, designing, and controlling systems. Moreover, multi-level (MRAP) multiple (MMRAP) are extensions derived from for practical modeling real-life problems. However, while formulating model, two problems mentioned above have some restrictions not deal with real world lost...