Tikrit Journal of Administrative and Economic Sciences

The transportation model is the most important linear programming model on which the economic transfer of the produced units is based from the sources of production (sources) to the demand sites (destinations) at the lowest possible transportation cost, and in the shortest possible time to meet the needs of those centers. The transportation model is considered complementary to the production process according to the production needs For this process, the transportation problem is one of the important mathematical methods that helped in the economic decision-making process for decision makers, and the importance of the transportation problem appeared when the production facilities faced economic problems when distributing goods from factories to the consumer. The research aims that the transportation problem is one of the vital and important problems to determine the lowest possible transportation cost for all goods, and given the lack of a specific method that gives an optimal solution except for the method of linear programming results when using the win Qsb program, so the researcher tried to study the most important ways to solve transportation problems that give an optimal solution Or a solution close to the optimum and compare the results of these methods with the result of linear programming to find out which method is identical or close to the result of linear programming, by applying it on the ground by taking one of the foodstuffs represented by the product (edible oil), which is transported from the company’s stores to three governorates (Erbil, Dohuk, Mosul) in order to reduce the total costs to a minimum. The study came out with the following conclusions. We conclude through the researcher's study of the transportation methods described above and their application on the ground. It has been shown through the extracted results that the best method is the method of linear programming results using the win Qsb program and the Vogel method that gives the optimal solution because it achieved the lowest possible total cost of transportation. Which was worth (6900) As for the critical path method, it needs a series of solutions in order to give the same result. Its value was (10400, 8300, 7100, 6900) in the case of data related to (edible oil) from the stores of the company (Altunsa) to the warehouses of the governorates, and it is better than all primary methods (The northwest corner method, a less expensive method) and these methods gave a very high cost compared to the previously mentioned methods. This method is considered one of the modern and very important methods for administrative and economic decision makers because it seeks to reduce transportation costs to the lowest possible extent.

In this paper, a new distribution of continuous distributions based on the Marshall-Olkin weibull family is introduced, so the proposed distribution will be obtained by fitting the family with an inverse Rayleigh distribution. Thus, the proposed distribution contains four parameters, of variably form and is called the Marshall-Olkin- weibull inverse Rayleigh distribution. The cumulative distribution function, the probability density function of the new distribution, is introduced. In addition, the most important statistical properties of the new distribution were obtained, for example, the quantity function, ordered statistics, where the method of greatest likelihood was used, in order to estimate the parameters of the new distribution(MLES), establish preference, and obtain high flexibility in dealing with different types of data, so it was Testing the new distribution on a set of real data, which represents the tensile strength of glass fibers with a diameter of 1.5 cm. Through the test, Notice, that the new distribution is more closely related to the data that the other distributions, using several statistical criteria

Variable selection is a very helpful procedure for improving prediction accuracy by finding the most important variables that related to the response variable. Poisson regression model has received much attention in several science fields for modeling count data. Invasive weed optimization algorithm (BHA) is one of the recently efficient proposed nature-inspired algorithms that can efficiently be employed for variable selection. In this work, BHA algorithm is proposed to perform variable selection for Poisson regression model. Extensive simulation studies and real data application are conducted to evaluate the performance of the proposed method in terms of prediction accuracy and variable selection criteria. The results proved the efficiency of our proposed methods and it outperforms other popular methods.

Financial Decisions involve many intangibles that need to be traded off, to do that, they must be measured alongside tangibles whose measurements must also be evaluated as to how well, they serve the objectives of the decision-maker. The most popular method for making multi-criteria decisions is the Analytic Hierarchy Process, which is referred to by the symbol (AHP), which helps decision-makers to choose the best solution from among several options and selection criteria. It is used in a wide range of fields in operations research, quality engineering, and others.
Through the (AHP) process, a hierarchy of decision elements is built using comparisons between each pair of elements expressed in a matrix form. The paired comparisons produce weighted vectors that measure the importance of the elements and criteria with each other. In this research, the statistical analysis system (SAS V.9.4) was employed to apply the (AHP) process to extract results and outputs that include measures of criteria, determine appropriate choices, and make decisions

Linear multicollinearity is considered one of the most significant problems in model building, as it arises due to the high correlation and linear dependence among multiple explanatory variables, negatively affecting the efficiency of the maximum likelihood estimator in estimating the parameters of the inverse Gaussian regression model, especially when the response variable is skewed in the positive direction and follows an inverse Gaussian distribution. This study aims to review shrinkage estimators used to address the problem of linear multicollinearity in the aforementioned model and propose a new estimator called Inverse Gaussian Dawoud and Kibria Estimator. The performance of the proposed estimator was compared to a number of estimators in a simulation study, where the results of Monte Carlo experiments showed that the proposed estimator contributes to reducing the mean square error of the model and achieves better performance compared to previously used estimators.

This study presented a new continuous distribution (Odd JCA Exponented Exponential Distribution) where this new distribution is an expansion of the Exponential distribution raised. Some mathematical and statistical properties of this distribution were studied, such as the function of danger, moments, mean, variance, the function that generates moments and goals, arranged statistics, some shapes, and the expansion of the probability density function that is used to find many properties. To demonstrate the flexibility of the new distribution, it was applied to a set of real data representing (66) plasma concentrations. for indomethacin. The proposed distribution was compared with eight other distributions, as it proved a high suitability for the data based on informatics standards (AIC, BIC, LL, CAIC, HQIC, A, W, PV, KS).

Structure equation modeling has gained wide popularity among most researchers, as it is an effective tool in analyzing and summarizing the results accurately in all research and studies, as it is considered an accurate tool in multivariate analysis. Variables, as it allows researchers to model latent variables and measurement errors simultaneously, and the M.L. This default method is used in most programs and the covariance matrix is calculated regardless of the nature of the data. Calculating the covariance matrix as inputs in order to improve the quality of the match and reduce the error. The structural model consists of four explanatory variables, and one response variable. Through the results of the study, the researcher found that the Kandal matrix is the best matrix as input for modeling through the conformity indicators represented (Quality Conformity Index, GFI, Standard Conformity Index, TLI, Comparative Conformity Index CFI). (If the values of these indicators were higher than the values of the same indicators when using Spearman’s matrix and the covariance matrix, as well as it was noted that (the root mean residual index SRMR, the root mean rounding error RMSEA) is less than their values in the rest of the matrices, which indicates the strength of the proposed matrix and it is preferable to use it as an input for modeling when using the M.L. The study was applied on realistic data consisting of 296 teachers of Nineveh education about the effectiveness of e-learning in primary schools, where the dependent variable in the study, technical and moral support, teacher characteristics and personal achievement were the explanatory variables. After analyzing the data, the study proved the effectiveness of e-learning. Accordingly, we recommend the General Directorate of Education in Nineveh and educational institutions to pay attention to the aspects whose paragraphs came in the questionnaire in order to obtain high quality and effective education.

In this paper we propose a new distribution with four parameters, called Linear Failure Rate of Alpha Power Distribution, this distribution is consists of the composition of Linear Failure Rate Distribution with Alpha Power Family, by adding the parameter of both models, so that we have a new distribution and can be used effectively in modeling survival data and reliability problems. This model has a batter feature or the probability distribution (PowLFR) it has a better feature and effectively than other distributions in the application to the real data. Statistical properties and we use the maximal likelihood to estimate parameter we derive and study some statistical properties and the proposed model; the practicability of the model is illustrated in the application on real data to show the fitting through some statistical criterion of gaudiness of fit.

The research aims at evaluating and selecting the supplier according to sustainability criteria by applying the process of fuzzy hierarchical analysis. The research was applied in Al-Zawraa State Company, and the research started from the problem of the company’s dependence on the principle of tenders in evaluating and selecting suppliers according to the conditions and controls set by the company, without taking into account the sustainability criteria. Al-Zawraa State Company for the manufacture of electrical equipment was chosen as a site for the application of the practical aspect. A sample of experts and specialists in the departments (commercial, technical, engineering, quality) were selected to survey their opinions on conducting bilateral comparisons between standards and suppliers. The researcher reached several conclusions, the most important of which is that Dar Al-Salaam is better equipped to meet the requirements of sustainability, and this is consistent with the opinion of the company despite the difference in the approved standards and method of evaluation.

In this paper we present a new distribution called Probability Distribution (Marshall Olkin Exponential Generalized- Inverse Weibull Distribution) derived from the merge of the Marshall Olken Family and generalized exponential distribution with the inclusion of Inverse Weibull distribution in the resulting formula from the combination of the two methods. Study the mathematical and statistical properties of this distribution such as; hazard function, Survival function, quantile function, Survival function normal moments, moment generation function, entropy and shapes and study the expansion of the function for ease the process of finding the properties. Use the greatest possible method in estimating the distribution parameters. A set of data is used to demonstrate the suitability and application of the new distribution. For the purpose of comparison, we use different models to test the quality and suitability of distribution

In this paper, we proposed a new distribution named Gompertz Inverse Weibull (GoIW) by using the Gompertz generalized family of distributions. The study extensively examined the significant properties of the GoIW distribution and utilized the maximum likelihood estimation technique to estimate its model parameters. A real data on rainfall was used to demonstrate the distribution's importance and compare it to commonly used lifetime distributions. The research results indicate that the GoIW distribution outperforms other popular distributions like Weibull Inverse Weibull (WeIW) and many others, in terms of fitting performance

This article also introduces the Gompertz-Topp-Leone inverse Rayleigh distribution and a one-parameter probability model, the statistical properties of which are thoroughly examined. Expressions for its moments as well as moment- producing, quantile, reliability, and hazard functions. Using the greatest likelihood technique, we estimate the GoTLIR distribution's parameters (MLE).
The model's hazard function has an inverted bathtub form, and we suggest the GoTLIR distribution's utility. The inverse Rayleigh distributions of Gompertz Topp–Leone have provided a generic formulation for the new distributions' density and distribution. Function Moments and hazard rate expressions have also been provided. Voda made the IR
distribution suggestion in (1972).
An application has been with a real lifetime data from engineering. The following data represent the tensile strength, of 74 carbon fibers tested under tension at gauge lengths of 20mm, available in Bader and Priest (1982)

In this paper we employment the OJCA family of distributions to introduce a new generalization for Rayleigh distribution named OJCA Rayleigh distribution the new distribution with two parameters denoted by (OJCARay). The purpose of study is to derive some of its statistical properties such as, expansion of probability density function, quntaile function, moments, incomplete moments, entropy. Finally, we use the maximum likelihood method to estimate the parameters for the new distribution

We propose a new distribution called [0,1] Truncated inverse Weibull inverse exponential distribution with three parameters, which extends the inverse exponential distribution. We derive some mathematical properties of the proposed model including expansion of PDF and the quantile function, moments, generating function, survival, hazard functions entropies. The method of maximum likelihood is used to estimate the parameters of the distribution. We illustrate its potentiality with applications to a real data set which show that the new model provides a better fit than other models considered

This paper will examine temperatures in Iraq and in five different cites (Baghdad, Mosul, Basra, Rutba and Sulaimaniyah) as they are geographically distributed throughout the Iraq country. Their effects on natural conditions by using statistical methods for climate to study the significant differences in the aforementioned regions in addition to study the stationary of the time series or not for temperatures and rainfall amounts and study the historical movements of climate change and activity, as well as the current debates surrounding global warming. And forecasting temperatures from the year (2018-2020) and on a monthly basis from the by using ARIMA models for each city and compare among them by nonparametric test Kruskal Wallis and by relying on the packages and functions of the R programming language. We will also examine the four data sets that we mentioned earlier, with attention to the severity and frequency of the problem of global warming and Desertification. We will also discuss the issue of modelling disasters statistically as it relates to increased risks and losses and the associated economic impact and how to reduce the problem of global warming. So that financial and governmental institutions can understand to reduce the scale of such disasters and develop proposed solutions.

The purpose of this research is to look at how to estimate parameters in multiple linear regression models. When multicollinearity and outliers are present in the model, existing techniques such as Ordinary Least Squares (OLS) estimate, Ridge Regression (RR), Ridge Least Absolute Value (RLAV), and Weighted Ridge Regression (WRID) are compared with a proposed new technique, Weighted Ridge Least Absolute Value (WRLAV). Ridge Least Absolute Value (RLAV) regression is a reliable alternative to the OLS, especially when the disturbances follow non-normal distributions with outliers. Using some test scenarios, the (WRLAV) estimator is based on (RLAV), which retains insensitive to inaccurate measurements and highly robust in the presence of these problems. The proposed estimate (WRLAV) is very robust to contaminated data for heavy tailed errors and shows good performance under various situations of errors and different degrees of multicollinearity using the Root Mean Square Error (RMSE) criteria for Simulation study and Standard Error (SE) for numerical example that was worked out when compared to other techniques of estimation

Chemometrics' quantitative structure-activity relationship (QSAR/QSPR) modelling is crucial for making predictions. Chemical data has the unique feature of having a large number of descriptors and a small number of chemical substances. For the QSAR/QSPR prediction to perform better, descriptor selection must be used. It is thought that choosing a descriptor is an NP-hard optimization problem. In the literature, many metaheuristic algorithms have been developed to address this issue. One of the most intriguing new advances in application to enhance the exploration and exploitation capabilities of the basic metaheuristics algorithms is the use of hybrid metaheuristics algorithms. In order to maximize descriptor selection, this work suggests a new hybrid firefly algorithm (FFA) and black hole algorithm (BHA) that combines the FFA and BHA's strengths. In order to increase the exploration and exploitation capabilities of the FFA method in QSAR/QSPR modelling with high prediction performance and quick computing time, we hybridize in an effort to minimize trapping at local optima. The experimental results, which were based on eight benchmark chemical datasets, show that the proposed hybrid method may achieve good prediction performance while decreasing the number of selected descriptors and speeding up computation