Flight schedule design and fleet assignment are the main sub problems of the airline schedule planning which have the most effect on the costs and profit of the airline. In this paper, integrated flight schedule design and fleet assignment problem is described and genetic algorithm has been developed to solve this problem. It has a number of constraints and multi-layer permutation chromosomes with variable length.So, creating the initial population randomly and use of customary operators of evolutionary ALGORITHMS will not be efficient since the probability of feasibility is very low. For this purpose, a new function based on loop concept to create an initial population and new crossover and mutation operators has been developed. A genetic algorithm has been used within the main loop to optimize the redirection of the passengers. Four models with different numbers of airports and fleets are created as an input for the problem which has been solved by two and three islands genetic ALGORITHMS. Results show that in each iteration of the main loop, feasible answers are obtained, and finally there was a proper improvement in the costs. In larger models, there is a better Improvement in the costs and more difference between two and three islands ALGORITHMS. Three islands mode results in a better solution within a longer time. The developed algorithm can successfully find feasible optimal solution and it can be used for high dimensional problems in which there is no possibility of finding the optimal solution by using conventional methods such as MILP.

Nowadays, with the increasing use of various discrete data acquisition methods such as drones and digital cameras, image processing has found wide application. However, video data alone cannot play a significant role in urban management decisions until they are transformed into statistical sequences. In this paper, a system for detecting the number of cars per unit length and time is presented. In this method, video data is converted into statistical sequences of traffic indicators. First, the images corresponding to each frame are modeled into background images based on the Gaussian mixture model, which are resistant to lighting changes. This operation is performed on a large number of frames to create a learned background image. In traditional traffic image processing methods, modeling the background image was not considered, and conversely, in the proposed method, this model is used to detect moving objects. Then, by comparing each input main frame with the learned background image, moving cars are detected. The information on the number of cars per unit length and time, which corresponds to the concepts of traffic volume and density, is used to estimate traffic flow. Based on the simulations performed and the comparison of the obtained results with other results from different studies, the high performance of the proposed method in car detection and accurate counting, considering proper background image training, is demonstrated. Moreover, this method can be used for processing low-quality images.

One of the new issues that have been raised in recent years is the hub network design problem. The hubs are collection and distribution centers that are used for the purpose of less connections and more of indirect than direct communications. They are interface facilities which are used as switch centers to collect and distribute flows in the network. They determine routes and organize traffic between source-destination in order to provide high performance and be more inexpensive. In the hub location problem, the aim is to find a suitable location for the hub and routes for sending information from a source to a destination, in order to reduce costs and gain desired purpose by multiple transfers between the hubs. In this paper, teaching and learning based optimization, particle swarm optimization and imperialist competitive algorithm were studied for locating optimally hubs and allocating nodes to the nearest located hub nodes. Experimental results show that optimal location for hubs by using cluster-based optimization algorithm (TLBO) successfully has been performed with extreme accuracy and precision.

Introduction: When a stream is partially obstructed by a bridge pier, the flow pattern around the pier is significantly changed.Changes in flow pattern are the cause scour around piers. Bridge pier scouring estimates, is an important parameterin the design of bridges because inattention to it may cause damage or reduce the life of the bridge [1-4]. The safeand economical design of bridge piers requires accurate prediction of the maximum scour depth around theirfoundations [5]….

Meta-heuristics have already received considerable attention in various engineering optimization fields. As one of the most rewarding tasks, eigenvalue optimization of truss structures is concerned in this study. In the proposed problem formulation the fundamental eigenvalue is to be maximized for a constant structural weight. The optimum is searched using Particle Swarm Optimization, PSO and its variant PSOPC with Passive Congregation as a recent meta-heuristic. In order to make further improvement an additional hybrid PSO with genetic algorithm is also proposed as PSOGA with the idea of taking benefit of various movement types in the search space. A number of benchmark examples are then treated by the ALGORITHMS. Consequently, PSOGA stood superior to the others in effectiveness giving the best results while PSOPC had more efficiency and the least fit ones belonged to the Standard PSO.

The reports show a rapid growth in the numbers of attacks to the information and communication systems. Also, we witness smarter behaviors from the attackers. Thus, to prevent our systems from these attackers, we need to create smarter intrusion detection systems. In this paper, a new INTELLIGENT intrusion detection system has been proposed using genetic ALGORITHMS. In this system, at first, the network connection features were ranked according to their importance in detecting attack using information theory measures. Then, the network traffic linear classifiers based on genetic ALGORITHMS have been designed. These classifiers were trained and tested using KDD99 data sets. A detection engine based on these classifiers was build and experimented. The experimental results showed a detection rate up to 92.94%. This engine can be used in real-time mode.

Compression can be done by lossy or lossless methods. The lossy methods have been used widely than the lossless compression. Although، many methods for image compression have been proposed yet، the methods using INTELLIGENT skipping proper to the visual models has not been considered in the literature. Image inpainting refers to the application of sophisticated ALGORITHMS to replace lost or corrupted parts of the data so that visual difference cannot be inferred from the reconstructed image. In this paper، first we review some of the image inpainting ALGORITHMS and some of the image compression techniques using the inpainting ALGORITHMS، we propose a new inpainting based image compression algorithm that can improve the compression rate considerably. Simulation results show that our proposed method has reasonable visual quality in comparison with the other proposed image compression ALGORITHMS.

In this paper, a new robust approach based on Least Square Support Vector Machine (LSSVM) as a proxy model is used for an automatic fractured reservoir history matching. The proxy model is made to model the history match objective function (mismatch values) based on the history data of the field. This model is then used to minimize the objective function through Particle Swarm Optimization (PSO) and Imperialist Competitive Algorithm (ICA). This procedure leads to matching of history of the field in which a set of reservoir parameters is used. The final sets of parameters are then applied for the full simulation model to validate the technique. The obtained results showed that due to high speed and need for little data sets, LSSVM is the best tool to build a proxy model. Also the comparison of PSO and ICA showed that PSO is less time-consuming and more effective.