Julian Scott Yeomans

Stochastic systems are often overwhelmed by incompatible performance requirements and inconsistent performance specifications that can be difficult to identify when supporting decision models must be constructed. Consequently, it is often advantageous to create a set of dissimilar options that afford distinctive approaches to the problem. These alternatives should satisfy the required system performance criteria and yet be maximally different from each other in their decision spaces. This paper describes a stochastic bicriteria procedure that can generate sets of maximally different alternatives. This stochastic algorithmic approach is both computationally efficient and simultaneously produces the prescribed number of maximally different solution alternatives in a single computational run of the procedure.

Salah H. Abid, Saad S. Mahmood, Yaseen A. Oraibi

The estimation some asymmetric dynamical maps by selecting an appropriate feed forward artificial neural network, transfer function and node weights, is the aim of our paper. Fast Fourier Transform (FFT) as transfer function is used. By means of the mean square error (MSE), the experimental results of proposed algorithm will compared empirically, with the results of the same network but with traditional transfer functions, Logsig and Tagsig. Different cases of the dynamical systems, Asymmetric Logistic, Asymmetric Logistic -Tent and Asymmetric Tent-Logistic are discussed. The performance of proposed algorithm is best from others in all cases from Both sides, speed and accuracy.

Obayomi Abraham Adesoji, Ogunrinde Roseline Bosede

The paper utilized the application of nonstandard methods to create suitable numerical schemes for some initial value problems arising from nonlinear second order oscillator equations. We have applied the technique of Non-local approximation at the grid points and a set of normalized denominator functions to renormalized the discretization functions. The resulting schemes have been used to simulate the original equations and they have been found to have the same monotonic properties as the original equation.

Mustapha Hamidi

This paper, taking Travelling Salesman Problem as our object, wishes to develop a constructive algorithm to prove P=NP. Our algorithm is described as this process: initially, constructing a convex polygon that includes all points and creating a centroid of this convex polygon, next, examine the position of each point compared to the centroid's position; then, repeating the previous step until all vertexes are included into the optimal tour. This paper has shown that our constructive algorithm can solve the Travelling Salesman Problem in polynomial time.

Gang Liu

We present a closed-form solution for convex Nonlinear Programming (NLP). It is closed-form solution if all the constraints are linear, quadratic, or homogeneous. It is strongly polynomial when applied to convex NLP. It gives exact optimal solution when applied to LP. The T-forward method moves forward inside the feasible region with a T-shape path toward the increasing direction of the objective function. Each T-forward move reduces the residual feasible region at least by half. The running time of T-forward method is better than the interior point method by a factor of O(N^2.5L).

Abdramane Sermé

This paper aims to provide a bound of the forward error of the extended iterative refinement or improvement algorithm used to find the solution to an ill conditioned linear system. We use the additive preconditioning for preconditioner of a smaller rank r and the Schur aggregation to reduce the computation of the solution to an ill conditioned linear system to the computation of the Schur aggregate S. We find S by computing W the solution of a matrix system using an extension of Wikinson iterative refinement algorithm. Some steps of the algorithm are computed error free and other steps are computed with errors that need to be evaluated to determine the accuracy of the algorithm. In this paper we will find the upper bound of the forward error of the algorithm and determine if its solution W can be considered accurate enough.

Mustafa Kamal, Shazia Zarrin, Arif-Ul-Islam

In Accelerated Life Testing, acceleration factor is known or there exists a mathematical model which specifies the life stress relationship. However, there are some situations in which neither acceleration factor is known nor such models exist or are very hard to assume. In such cases partially accelerated life tests are better criterion to use. This paper deals with simple Constant Stress Partially Accelerated life test using type-I censoring. Assuming that the lifetimes of test item follow Inverted Weibull distribution, the maximum likelihood estimation procedure is used to obtain the estimates for the distribution parameters and acceleration factor. Asymptotic confidence interval estimates of the model parameters are also evaluated by using Fisher information matrix. A Simulation study is used to illustrate the statistical properties of the parameters and the confidence intervals.

T. Ganesh, PRS Reddy

In Managerial Decision making, the problem environment will be encircled by a set of alternatives for set of criteria. The main objective is to choose the best alternative under each criterion. In this contest, the Decision Maker (DM) plays an important role in solving the hard/complex problems. This type of scenario gives raise to the concept of MCDA. In this paper, we made an attempt to provide some algorithms which are user-friendly. In this paper, we have provided some algorithms which supports in computing the concordance and discordance indices.

Ch. R. Phani Kumar, P. Krishna Mohan, Ch. Satyavathi

Data processing through cellular networks has been increased in the present decades. Huge amount of data can cause interruption to the cellular users. Hence a code has been generated that acts as a filter bank. Placing the code into the controlling element of the filter banks required selection of the channels for allocating the voice and data channels are made. By allocating the free channels to the calls, the user can able to provide service to all the calls and also to the data when any channel is free. Even the feedback for the lost data is also provided in such a way that all the channels in one or the other case are kept busy either by voice or data transmissions. Hence the code can provide effective utilization of the channels during data transmission.

Ch. R. Phani Kumar, B. Uday Kumar, V. Malleswara Rao, Dsvgk Kaladhar

Mobile network analysis has a huge potential that provide insight into the relational dynamics of individuals. Machine learning and data mining techniques provide the behavior patterns of the mobile network data. The data transfer during all the days has produced good results in transfer of data starting from Day 1 to Day 22. Hierarchical clustering of the datasets for all the 1634 data examples in the mobile traffic dataset. Complete linkage dendrogram has been produced between 0 and 4.64. Two clusters have been produced from the present wireless mobile traffic datasets.