Articles In Press

    Transportation problems is a wide-ranging class of optimization models in operations research that address the problem of optimal allocation of a large number of items from multiple origins to multiple destinations on several conflicting objective functions. It studies the sensitivity of multi-objective, multi–item transportation problems and analyses how optimal solutions are affected by different parametric changes. Here we propose an integrated optimization model for resource consumption over time and cost. WinQSP software was used to demonstrate an application of a three-supplier, four-demand center, two product type system. Transport costs are the most sensitive parameter (coefficient of variation, 0.85-1.35) followed by supply constraints (0.90-1.25) and demand fluctuations (0.95-1.20), the results show Simulation results indicate that, until a certain parameter limit (±10%) there is no significant change to the optimal solution. Beyond these limits parameters will begin to significantly affect their respective solutions. The sensitivity analysis emphasizes the importance of monitoring certain parameters closely in practice. The current research adds to the literature by providing a quantitative measure of robustness and stability in optimal solutions for multi-objective transportation systems that are instrumental for decision-makers.

Evaluating the mechanical response of a cylindrical concrete is crucial for structural reliability. This paper develops a multi-objective optimization framework to analyze concrete performance under uniaxial compression and split tensile loading, utilizing the Von Mises stress (S, Mises) and Plastic strain (PE) to model the failure of concrete. The parameter estimation is cast as a cubic programming problem to handle the non-linear nature of concrete degradation. To resolve competing structural objectives, a multi-objective model is established, and the resulting Pareto front is solved via the harmonic mean method—an approach that prevents outlier distortion and balances tensile and compressive capacities. The numerical findings confirm that this methodology significantly enhances predictive accuracy over traditional single-objective and arithmetic mean approaches, providing a stable, computationally efficient solution for optimizing concrete mixes and predicting failure states.

Predicting precipitation amounts is a complex problem due to the non-linear nature of the precipitation processes, the significant interactions between factors influencing precipitation formation, and the varying levels of uncertainty resulting from climatic variability. This study aims to develop a model based on Bayesian Neural Networks (BNN) to predict monthly precipitation amounts and represent the uncertainty inherent in these predictions. The model applied to climatic data for a region from north of Baghdad to the Iraqi-Turkish border for the period 1984–2024, obtained from the NASA POWER database. The data included a set of influencing climatic variables: specific humidity at a height of 2 meters (QV2M), temperature at a height of 2 meters (T2M), wind speed at a height of 2 meters (WS2M), and surface soil moisture (GWETTOP), in addition to monthly precipitation amounts as the target variable. Because precipitation is influenced by a range of interacting and simultaneous factors, several interaction variables were derived to enhance the model's capabilities and improve its predictive flexibility. The results demonstrated the proposed Bayesian model's ability to accurately represent temporal precipitation patterns and to handle the uncertain and irregular nature of the studied climatic phenomenon. The model achieved a coefficient of determination (R²) of 0.821, which demonstrates its ability to explain a large proportion of the variance in monthly precipitation amounts. Ablation analysis performed to interpret the model's performance and determine the importance of the variables. The results showed that the Moisture_Thermal_Index variable had the greatest impact on the model's performance, followed by Evapo_Proxy, then GWETTOP and GWETTOP_lag1. This underscores the importance of interactions between moisture, temperature, evaporation processes, and soil moisture status in explaining precipitation dynamics. The findings of this study confirm that integrating BNN with physical knowledge-based climate variable engineering provides an effective methodology for precipitation forecasting. This framework can serve as a supportive tool for decision-making in water resource management, environmental planning, and climate change risk assessment. 

With the advancement of technology, the increasing size of software, and the need to develop it in record time and with less cost and effort, the need arose for a method to reuse software components and integrate them into other similar programs, Researchers have conducted reviews on the use of artificial intelligence, machine learning models

This review discusses the most important software metrics used as inputs in prediction models, . It also shows how machine learning techniques and artificial neural networks have been used instead of traditional statistical methods.

 This review highlights the challenges faced by researchers in this field.

    Software requirements are a determinant of whether technical and software projects will be successful or not. Nonetheless, the fact that they are dependent on natural language makes them susceptible to basic flaws, the most significant of which are ambiguity, inconsistency, and misclassification. Considering the slowness and human bias of traditional methods and manual reviews, and the fact that current automated tools tend to do the same tasks in a disjointed and isolated way, there is an immediate necessity of more comprehensive and smart way of doing it. The review, in turn, recognizes a major methodological paradigm change through evaluation 

Diabetes related foot ulcers and the associated general complications of diabetes pose significant health issues around the globe. Skilled professionals and specific medical apparatus are typically required by a number of diagnostic approaches for wound assessment. The use of artificial intelligence in the classification of diabetic foot ulcers is examined in this study specifically looking at machine learning and deep learning methods. The research shows that computers running on artificial intelligence improved diagnoses for foot ulcers are made. Artificial intelligence techniques are currently changing the way DFU management is carried out by means of an improvement in  diagnostic accuracy.

 In this research, the phenomenon of evaporation was studied for the period     1980 - 2020 for the city of Mosul, and the adaptive filtering method and neural networks were used in order to predict the monthly evaporation values in the city of Mosul. if the adaptive filtering is characterized by correcting the parameters of the time series model to the extent that the parameters that give the lowest mean of the error squares MSE for the purpose of predicting the future values of the phenomenon under study. The NARNET network was used as a neural network to compare the two methods, as statistical criteria were adopted to compare them and identify the best method for prediction.

A factorial seasonal model was diagnosed and its parameters were found using the computer programs Excel, Minitab and MATLAB, and future values were predicted. Through the comparison, it was found that the neural networks represented by the NARNET network give relatively better results compared to the conditioned purification.

The research aims to study the effectiveness of wave diagnostics for single-variable time series models, by comparing a number of traditional methods based on Wave analysis, random data of various sizes (Small, Medium, Large) and multiple iterations were generated through data with specific specifications for the selection of hybrid models that combine simulation and Wave analysis, as well as the traditional exponential Bootstrap was compared with the hybrid model with waves through some statistical criteria.As well as comparing the performance of the traditional exponential bootstrap model with a hybrid model that combines seasonal exponential Bootstrap and the transformation of wave filters in the analysis and prediction of time series, through the application to the two models of a set of data with seasonal characteristics and high oscillations, the results showed that the hybrid model achieves higher accuracy in forecasting, and is characterized by its ability to isolate noise and sudden changes an effective option to improve the quality of forecasting, especially in contexts characterized by Stability or complex seasonality.

The conjugate gradient method is an essential category of approaches for solving unconstrained optimization problems. In this paper, we are considering five variables that are related to the mechanical properties of steel and concrete for a pre-stressed concrete beam. The aim is to optimize two outputs which are maximum principal strain and maximum deflection of the structure.  Two surrogate models are constructed using design sampling method for preparing the involving models. The reliability of the surrogate models is checked by the coefficient of determination ( ). We are using unconstrained nonlinear optimization by applying conjugate gradient method with Polak and Ribiere parameter. Five optimal iterations are executed in the presented algorithm for each model with substitution in each other. Optimized value of each variable in addition to the optimum two surrogate model outputs were determined and verified with the variable range values and exact model outputs. Finally, sensitivity analysis was performed and the main rational percentage effect of each variable on the outputs and their interactions were determined.  

Lifetime data analysis is essential in survival analysis and reliability engineering. Traditional distributions like the exponential and Weibull often fail to capture skewness, heavy tails, or non-monotonic hazard rates. This review examines transformation-based methods that enhance baseline distributions by introducing additional shape parameters, improving flexibility for complex patterns such as bathtub-shaped and multimodal hazards. These techniques provide parsimonious yet adaptable models suited to diverse real-world phenomena. A comparative evaluation highlights each method's strengths and applications, guiding appropriate selection based on data characteristics. This study serves as a practical reference for researchers modeling lifetime data across biomedical, engineering, and other disciplines.

Inflation, with its abnormal rates, is one of the most obvious negative economic phenomena affecting most economies worldwide due to its effects on various economic, social, and other spheres. Thus, the purpose of this study is to examine the relationship between Iraqi inflation rates and oil prices for the period from December 2010 to January 2024 using the autoregressive vector model, a multivariate time series model that considers the interaction between variables and treats all study variables as end-variables. Based on shock analysis and causality testing, the results demonstrated a unidirectional causal relationship between oil prices and the rate of inflation.

This paper develops a hybrid statistical framework using Robust Partial Least Squares Regression (RPLSR) with wavelet analysis to increase model accuracy and stability under noisy and high-dimensional conditions. Classical Partial Least Squares Regression (PLSR) is famous for being sensitive to multicollinearity but also sensitive to outliers and data irregularities. The proposed solution addresses these limitations, with robust estimation procedures and wavelet-based denoising from Daubechies, Coiflets, and Discrete Meyer wavelets. Several sample sizes and predictors were used in the simulation to determine performance by MSE and R2 through the means of varying sampling size and predictor size of predictors. The results showed that all wavelet-enhanced RPLSR variants significantly improve estimation accuracy and increase predictive power, with an R2 of over 97% in some cases. The model is further validated by real data from Al-Zahraa Hospital in Iraq, as well as full blood count observations in complete blood samples at the Al-Zahraa Hospital, where the methodology was tested for robustness and interpretability. In general, robust regression with wavelet preprocessing provides a powerful, flexible, and noise-resilient framework for analyzing complex biomedical, environmental, and financial datasets.

The multivariable individual observations chart is based only on values of the views, and it does not account for changes that may occur in production. Thus, the researchers proposed producing new charts that could control and monitor equivalence and variance of multivariate individual observations using discrete wavelet transforms of approximation and detail coefficients for Daubechies wavelets across orders to measure production by monitoring the development process. The target line in the approximation and detail coefficients chart was estimated to be an average of the approximation and detail coefficients at the mean of the approximation and detail coefficients curve. In both cases, the upper and lower control limits are the target line added (or subtracted) by three multiplied by the square root of the factor in the variance-covariance matrix between the approximation and detail coefficients. The simulation and real data were used to make the proposed charts in the MATLAB program intended for this purpose. The findings of this study presented the effectiveness of the proposed charts and the possibility of using them to control and monitor the quality of the production process.

In this article, we proposed three classes of multiphase multivariate mixture ratio estimators for estimating the population mean with multi-auxiliary characteristics in partial information case (PIC) situation. The suggested estimators (PIC-I, PIC-II and PIC-III) combine data from many auxiliary variables and auxiliary attributes to reduce the mean square error (MSE) to go beyond the existing single-variable approach. The analytical expressions for the MSE of the estimators are derived up to the first order of approximation. The efficiency of the proposed estimators is theoretically compared with the existing ratio-type partial information case estimator. Empirical results based on simulated data demonstrated that the proposed multivariate mixture estimators in partial information case situations (PIC-I, PIC-II and PIC-III) were consistently more efficient than the existing partial information case estimator.

In this work, we present a comprehensive review of machine learning, a vital branch of artificial intelligence that evolves rapidly across variety of scientific fields. We address the basic concept of machine learning, explore the peculiar characteristics of machine learning, and review its main types.

Furthermore, we highlight various applications of machine learning within different fields, such as security, natural language processing, agriculture, and healthcare. Number of important studies with high contribution to this field have been discussed in this review, considering their findings and future paths in the field, and focusing on recently emerged innovative methodologies.

this approach relies on the methodical integration of previous beliefs (expressed through initial distributions) with fresh data obtained from real-world experiments (using a probability function). In sophisticated computations, Bayesian statistics presents difficulties despite its many benefits. However, these difficulties have been largely overcome because to developments in simulation methods like the MCMC. The MH algorithm, Gs, MCH, and NUTS technique are the four main algorithms that are examined in this paper within the MCMC framework. The study compares the benefits of the Bayesian approach, such as model design flexibility and uncertainty management, with its drawbacks.