Sriyankar Acharyya

Ant Colony Optimization Technique Applied in Network Routing Problem

In this paper, the Ant Colony Optimization Technique has been applied in different network models with different number of nodes and structure to find the shortest path with optimum throughput. Three variations of the Ant Colony Optimization Technique, ACO1, ACO2 and ACO3 has been proposed and applied on different standard network models and the results has been analyzed and concluded. A Tabu list is also maintained for a network with large number of nodes and results were collected to find the optimum size of the Tabu list in one of the algorithms proposed here. Experiments have also been performed by varying the load of the network. Here the throughput and the reliability of the network has been specially taken as the performance factor of the network.

Optimal task scheduling in cloud computing environment: Meta heuristic approaches

Cloud computing is the latest continuation of parallel computing, distributed computing and grid computing. In this system, user can make use of different services like storage, servers and other applications. Cloud resources are not only used by numerous users but are also dynamically redistributed on demand. Requested services are delivered to users computers and devices through the Internet. The fundamental issue in cloud computing system is related to task scheduling where a scheduler finds an optimal solution in cost-effective manner. Task scheduling issue is mainly focus on to find the best or optimal resources in order to minimize the total processing time of Virtual Machines (VMs). Cloud task scheduling is an NP-hard problem. The focus is on increasing the efficient use of the shared resources. A number of meta-heuristic algorithms have been implemented to solve this issue. In this work three meta-heuristic techniques such as Simulated Annealing, Firefly Algorithm and Cuckoo Search Algorithm have been implemented to find an optimal solution. The main goal of these algorithms is to minimize the overall processing time of the VMs which execute a set of tasks. The experimental result shows that Firefly Algorithm (FFA) performs better than Simulated Annealing and Cuckoo Search Algorithm.

A SAT approach for solving the nurse scheduling problem

Nurse scheduling problem (NSP) represents a subclass of constraint satisfaction problems (CSP), involving a set of constraints. The problem is highly constrained and difficult to solve. The goal is to find high quality shift assignments to nurses satisfying constraints related to labor contract rules, requirements of nurses as well as the employers in health-care institutions. The constraints are classified as hard and soft, depending on their importance. In this paper, a real case of a cyclic nurse rostering problem is introduced. dasiaCyclicpsila means that the generated roster can be repeated indefinitely if no further constraint is introduced. In earlier investigation we saw that simulated annealing performed better than other local search techniques. In this paper we have converted NSP to a satisfiability problem(SAT) and applied GSAT to solve it. We show that GSAT incorporated with a tabu list has outperformed other methods, like, simulated annealing and genetic algorithm in almost all instances.

Meta-heuristic approaches for solving Resource Constrained Project Scheduling Problem: A Comparative study

Meta-heuristics for solving Combinatorial Optimization Problems (COP) is a rapidly growing field of research. In this paper we have considered the Resource Constrained Project Scheduling Problem as a COP. The problem is highly constrained and is a common problem for many construction projects. The problem is NP-hard and deterministic methods are slow in execution. In our work, we use Simulated Annealing, Tabu Search, Genetic Algorithm, Particle Swarm Optimization and Elite Particle Swarm Optimization with Mutation for solving benchmark instances of this problem and compare their performances with each other. The results show that Simulated Annealing outperforms other methods in getting optimal results with minimum number of fluctuations.

Hybrid Local Search Methods in Solving Resource Constrained Project Scheduling Problem

Now-a-days different meta-heuristic approaches, their variants and hybrids are being applied for solving Combinatorial Optimization Problems (COP). In this paper Resource Constrained Project Scheduling Problem (RCPSP) has been presented as a COP. This is a common problem for many construction projects. It is highly constrained and is categorized as a NP-hard problem. In our earlier work Simulated Annealing (SA_RCP) outperformed other meta-heuristics, like, Genetic Algorithm, Tabu Search, Particle Swarm Optimization and its variant in solving benchmark instances of this problem. Having been inspired by this result we have further developed new hybrids of Simulated Annealing and Tabu Search. In this work, we have proposed five more methods developed by combining Simulated Annealing and Tabu Search and applied them for solving a benchmark instance of this problem. The results show that Simulated Annealing incorporated with Tabu List, Greedy Selection Heuristic and aspiration criteria (GTSA_AC_RCP) outperforms other methods in getting optimal results with maximum hit and minimum fluctuations.

Neural model of gene regulatory network: a survey on supportive meta-heuristics

Gene regulatory network (GRN) is produced as a result of regulatory interactions between different genes through their coded proteins in cellular context. Having immense importance in disease detection and drug finding, GRN has been modelled through various mathematical and computational schemes and reported in survey articles. Neural and neuro-fuzzy models have been the focus of attraction in bioinformatics. Predominant use of meta-heuristic algorithms in training neural models has proved its excellence. Considering these facts, this paper is organized to survey neural modelling schemes of GRN and the efficacy of meta-heuristic algorithms towards parameter learning (i.e. weighting connections) within the model. This survey paper renders two different structure-related approaches to infer GRN which are global structure approach and substructure approach. It also describes two neural modelling schemes, such as artificial neural network/recurrent neural network based modelling and neuro-fuzzy modelling. The meta-heuristic algorithms applied so far to learn the structure and parameters of neutrally modelled GRN have been reviewed here.

ACO variants in solving Mobile Ad hoc Network routing

Mobile Ad hoc Network (MANET) is one of the most upcoming technologies in the field of wireless communication. As there is no predefined network configuration in MANET, routing, therefore, is much more difficult compared to that in a wired network. It is known that in MANET there is a continuous change in topology resulting in the change in adjacent nodes to a certain node. This subsequently may play a vital role in finding the path from source to the destination host. Meta-heuristics, like Ant Colony Optimization (ACO) are being used successfully in MANET routing. In our earlier work we made some variations of ACO techniques for routing in wired network. Here, in this paper, some modifications on those variants of ACO have been made considering the situations in MANET. Variations are related to selecting the next node to visit depending on the number of adjacent nodes to the current node and modification of the pheromone deposit formula on the basis of transmission time. In some variants a tabu list has been incorporated to enhance the performance remarkably. Using NS2 simulator the experiments have been made to compare the performance of six variants of ACO, where the various parameters have been adjusted optimally. It is observed that modified versions are better than the previous ACO variants in overall performance related to throughput and packet drop rate.

Constraint Satisfaction Methods for Solving the Staff Transfer Problem

The Staff Transfer Problem is concerned with the assignment of transfer postings to employees in large organizations. Staff transfers are an important issue in Human Resource Management in countries like India and China that have many large public sector undertakings. The Staff Transfer Problem can be viewed as a Constraint Satisfaction Problem, and methods such as Simulated Annealing, Genetic Algorithms, Satisfiability (GSAT), and Conflict Directed Backjumping can all be employed to solve randomly generated problem instances. Computer experiments indicate that Simulated Annealing is the best method of solution. GSAT with a tabu list yields solutions of good quality but is unable to solve large instances. Genetic Algorithms is also good, but it takes much more time than Simulated Annealing, Conflict Directed Backjumping, a deterministic search technique, is markedly inferior to the other methods. Thus for solving the Staff Transfer Problem, randomized approaches appear to be superior to deterministic ones.

Gene expression profiling by estimating parameters of gene regulatory network using meta-heuristics: A comparative study

Genes are present in the nucleus of every cell in an organism. Genes, metabolites, proteins and other by-products of cellular activity form a signaling pathway or network which is called a Gene Regulatory Network. Computational reconstruction of the network may uncover potential genetic causes of diseases and may aid drug detection. Advancements in biotechnology and image processing tools have made time series gene expression data available to researchers of computational biology. Reconstruction of Gene Regulatory Network has found a new direction with the availability of this data. After being processed by different statistical methods, the time series data may be considered as a matrix with each row representing a gene and each column representing a time point. The data suffers from an insufficiency of number of columns in relation to number of rows. This makes the reconstruction process more tedious. The problem is known as Curse of Dimensionality problem. The methods which are described here take processed microarray gene expression data as the input and produce the simulated gene expression time series with larger number of columns having regular small intervals. Gene Regulatory Network is reconstructed in the framework of Recurrent Neural Network. The parameters of the network are iteratively optimized using efficient local search optimization algorithms, namely two variants of Simulated Annealing and Tabu Search. The optimized parameters are used for the comparative study between the three methods in producing the time behavior or expression profiles of the genes. For almost all genes, the simulated profiles closely correspond to the original profiles.

Parameter Estimation of Gene Regulatory Network Using Honey Bee Mating Optimization

Computational biology is the esteemed interdisciplinary field where expertise from the fields like Mathematics, Statistics and Computer Science are applied to have the insight in biological phenomenon. Advanced methods and techniques of biotechnology and allied fields facilitate the availability of biological data to the researchers of computational biology. Microarray time series gene expression data is such an effective dataset which uncovers the regulatory relationships between any pair of genes in a gene set and hence facilitates the reconstruction of Gene Regulatory Network. The Artificial Neural Network environment is used to find the expression level of a gene at time t+?t in terms of the available expression level at time t. The underlying network parameters are uncovered as the simulated time series are compared with available real dataset in successive iterations. Estimation of the parameters of gene regulatory network is an important research area to be addressed. Here in this paper, the parameters are estimated using Honey Bee Mating Optimization algorithm. The intelligence of queen bees of the bee colony to select prospective drones for mating, crossover and mutation to support effective new genotypes and nurture of the good broods by worker bees is applied to solve the optimization problem of Parameter Estimation. Two experiments are conducted here. In experiment 1, the simulation based on the synthetic dataset of predefined parameters showed good performance accuracy. In the case of experiment 2, where real dataset was used, the cost convergence indicates the excellence of Honey Bee Mating Optimization in Parameter Estimation of Gene Regulatory Network.