In this paper, we focus on finding an optimal approximate solution for a specific type of optimization problem known as the fuzzy transportation problem, using pentagonal fuzzy number. In this context, the cost, supply and demand values associated with fuzzy transportation problems are represented as pentagonal fuzzy numbers. We convert these fuzzy numbers into crisp values using the range technique and solve the problem using the max-min method, which is applicable to transportation issues. The feasibility of our proposed method is demonstrated through numerical examples and the results are compared with those obtained from existing methods.
Since our approach directly extends the classical method, it is straightforward to understand and can be easily applied by decision-makers to real-life transportation problems.

KEYWORDS: Optimization; Cost-Efficient; Transportation Problem; Pentagonal Fuzzy Number; Range-max-min Method.

The objective of this work is to propose a topological method for predictive modeling of machine learning, which is considered a technique serving Data Science, essential for data modeling. The Topological Discriminant Analysis (TDA) proposed is according to the type of data, quantitative, qualitative or mixed explanatory variables. This decisional topological classification is a supervised clustering method attempts to discover the intrinsic structures and discriminant information embedded in the data. There are many predictive techniques, they are most often and most usefully applied to various problems in many fields.

Classification is therefore the operation which allows each individual of the population studied to be placed in a class, among several predefined classes, according to the characteristics of the individual indicated as explanatory variables. The proposed topological approach of discrimination is based on the notion of neighborhood graphs in a decisional context.

The explanatory variables are more or less correlated or linked depending on whether the variables type, quantitative, qualitative or a mixture of both. This topological model of discrimination analyzes the structure of the correlations or dependencies observed in each class according to the explanatory variables.

To validate the effectiveness of our topological approach, a series of experiments are performed on several UCI benchmark datasets, with quantitative, qualitative and mixed explanatory variables. The results are compared to those of different existing predictive modeling techniques of machine learning.

KEYWORDS: proximity measure; neighborhood graph; adjacency matrix; discriminant analysis; predictive modeling.

The Buckley-James estimator (BJE) is an estimator of ? for the semi-parametric linear regression model Y = ??X + W with right-censored data. Several iterative algorithms for the BJE have been proposed so far. However, they may either converge to a value far away from the BJE, or fail to converge at all. On the other hand, Yu and Wong (2002) introduced a non-iterative algorithm for finding all solutions of the BJE. While theoretically appealing, this approach becomes computationally intensive if ? ? Rp with p > 1 and with a large sample size n. This paper presents a modification to the BJE with a non-iterative algorithm. It yields the exact BJE if
? ? R; otherwise it consistently approximates both ˆ? and the true parameter ? as n increases. We compare its performance against the BJE through simulation studies and illustrative examples. We also carry out a data analysis of a real-world data set.

KEYWORDS: Linear regression; survival analysis; right censorship model; semiparametric model; Buckley-James Estimate.

In this paper, a one-parameter lifetime distribution called the Tal-Ajah distribution for modeling lifetime data is proposed. Mathematical and statistical properties of the new distribution, including its survival function, hazard rate function, shape characteristics of the density, stochastic ordering, entropy measure, and stress-strength reliability, are studied. Two datasets, including a process dataset and a 40-blood cancer patient dataset, are used for the comparative study. The estimation of parameters is carried out using the method of maximum likelihood. The goodness of fit of the model is assessed using HQIC, BIC, CAIC, and AIC. The proposed distribution is compared with the Exponential, Lindley, Ishita, Akash, Pranav, Christ-Jerry, Shanker, and Rama distributions, and it shows superiority over the competing models.

KEYWORDS: Tal-Ajah distribution; Moments; Goodness of fit; Hazard rate function; StochasticOrdering.

Tropical cyclones emerging over the Arabian Sea and Bay of Bengal pose significant threats to coastal India, yet their meteorological characteristics and impacts differ due to contrasting oceanic and atmospheric conditions. This study presents a comparative analysis of two major cyclonic systems that struck India in 2023, cyclone Biparjoy (Arabian Sea) and Cyclone Michaung (Bay of Bengal). A multidisciplinary framework  integrating statistical tools, circular distribution modeling, and GIS-based spatial analysis was employed to assess cyclone dynamics, wind behavior, pressure variations, rainfall intensity, and resultant flooding impacts. The study applies Von Mises and Wrapped Normal distributions to model storm directions, Mann-Whitney and regression tests to examine meteorological relationships, and spatial interpolation techniques to visualize precipitation and wind intensity. Results reveal that while Biparjoy exhibited a prolonged lifespan and north-northeast trajectory with moderate inland rainfall, Michaung was shorter-lived but more destructive in urban regions, causing severe flooding in Chennai due to extreme precipitation (530 mm in 72 hours). Comparative findings underscore how regional atmospheric conditions and coastal geomorphology govern cyclone intensity and flood susceptibility across India’s east and west coasts. The integrated analysis provides valuable insights for enhancing cyclone forecasting, flood preparedness, and adaptive disaster management in both basins.

KEYWORDS: Sections; Tropical cyclones; Von Mises; Wrapped Normal Distribution.