نشریه تصمیم گیری و تحقیق در عملیات
سال پنجم شماره 2 (تابستان 1399)

The Cramer-Rao lower bound is obtained by using integration by parts and the Cauchy-Schwarz inequality.  The integration by parts formulas of Malliavin calculus plays a role in this study. The point estimation problem is very crucial and has a wide range of applications. When we deal with some concepts such as random variables, the parameters of interest and estimates may be observed as imprecise. Therefore, the theory of fuzzy sets is important in formulating such situations. Using the fuzzy set theory, we define a fuzzy-valued random variable and fuzzy stochastic process.  We use the Malliavin derivative and Skorohod integral to study the asymptotic properties of the statistical model for fuzzy random variables. We show how to use the conditional expectations of certain expressions to derive Cramer-Rao lower bounds for Fuzzy valued Random Variables that they do not require the explicit expression of the likelihood function. As an example, we consider a fuzzy random sample of size n induced by independent standard normally distributed random variables with fuzzy parameter.

Assessing the reliability and availability of the production system reduces the likelihood of sudden and costly stops, which is very risky. To this end, this paper attempts to provide a new way to determine the reliability and availability of a production system, which can be used for a variety of failure components such as materials, supplies, personnel and machinery. Therefore, using Fuzzy Bayesian Approach, unrealistic events that are actually created in a production system have been processed and the proposed model has been used to assess the condition of Pegah Fars milk factory, so that the rate of failure and repair and reliability of components and system with Bayesian method was calculated. And because the available information space is uncertain, the reliability parameters became fuzzy. Then the availability of the components and then the whole system was calculated using the formula provided by Martz and Waller and the Bayesian method, and the parameters of availability were converted into fuzzy. Finally, the information obtained about the reliability and availability of the system and components were analyzed, that the results show the improved approach, provides a more accurate estimate of reliability and availability.

Today, organizations need to develop their production systems to achieve a competitive advantage and maintain it in the market. To achieve this, a completely new and innovative approach is needed to manage and produce an organization that dramatically increases productivity and also helps create effective and efficient supply chains. Now that the industrial revolution has grown to the fourth generation, organizations can use its new technologies to achieve and maintain a competitive advantage, as well as to increase productivity. The present study examines the interactive relationships of fourth-generation industry technologies in selected food industries to achieve this; first, through literature extraction, 20 cases of fourth-generation industry technologies were extracted, and in the second stage, with the fuzzy Delphi approach, 13 cases of fourth-generation industry technologies were identified for an explanation in the food industry. Then, the R-DEMATEL was used to investigate the interactive relationships between fourth-generation industry technologies.

The purpose of this study is to identify the components and develop a model to provide rules for optimizing viral marketing in businesses. It is an applied research and in terms of method, it is mixed (quantitative and qualitative). The statistical population of the research in the qualitative part includes 15 people in the three generations X, Y and Z (Millennium marketing generation) and in the quantitative part includes 460 online buyers. Data collection tools were used in the qualitative part of projection technique and in-depth interview. Interviews were analyzed and summarized using MAXQDA software, through which six components were identified, and then in a small part of 12 experts were used to determine the index of CVR, and then exploratory factor analysis was performed by SPSS software. Because selecting the most effective new components of viral marketing can have a huge impact on the accuracy of the viral marketing model in online businesses, To identify the most effective components, genetic metaheuristic algorithm was used, which is the software used in this section, WEKA and RAPIDMINER. Finally, the rules of viral marketing optimization were identified using the decision tree method. Findings in the qualitative section indicate that online persuasion, online trust, online support, online services, online attractiveness and online risk-taking are components of viral marketing. In the quantitative section and genetic algorithm, it was shown that the online risk component could not be used as an effective component for modeling and extracting viral marketing rules, so it was removed from the six components.

Many of the world's top universities have already decided to hold the next semester with e-education. In our country, the forecasts show a red situation for some areas in terms of the prevalence of corona. Also, a number of university students are living in these areas. Therefore, in the planning of the next semester, more attention should be paid to e-education. It is recommended that the next semester be implemented in two parts, including e-education for 10 weeks and face-to-face training for 3 weeks. In face-to-face training with the needs of the educational and dormitory space, students are divided into two sub-categories (A) and (B) so that it is possible to implement health protocols in universities and dormitories. In this paper, to determine the number of male and female students accommodated in dormitories for both subcategories (A) and (B) in face-to-face training, a mathematical optimization model is proposed in the form of nonlinear programming with integer decision variables. In the objective function of the model, the distribution of students in the educational space and dormitories is done in such a way that it has the maximum possible dispersion (minimum difference), due to the prevention of the spread of corona disease. This model has been implemented for the allocation of student dormitories at Allameh Tabatabai University, and its use can bring positive results for decision makers.

In this paper, we present a new trust region method for unconstrained optimization problems with locally Lipschitz continuous, nonconvex functions. In this method, in the ratio test, the current objective function value is replaced with maximum of some objective function values in the previous iterations. The new method nonmonotone properties and prevents falling into narrow valleys. Proving global convergence requires only two conditions: 1- there should be is a sufficient reduction for the approximate model in the solution of trust region subproblem, 2- the approximation Hessian matrix be bounded. Then, the convergence property of this method is investigated. Finally, the presented method is implemented on some nonconvex problems in MATLAB environment and numerical results are compared with the nonsmooth trust region method.

Research has revealed that Data Envelopment Analysis (DEA) is an excellent method of data-‎based performance analysis for comparing decision-making units with multiple inputs and ‎outputs. Selecting inputs and outputs (performance measures) in DEA is a delicate task. In ‎principle, including a large number of inputs and outputs is a positive advantage. However, the ‎inclusion of multiple inputs and outputs might translate into a great deal more of additional ‎data being included, and this may lead to some decision-making units being considered and ‎designated as efficient simply because of their high performance in relation to a number of ‎redundant and useless variables. Elsewhere, in some situations, some performance measures ‎can play both an input and output role. These performance measures are called flexible ‎measures or dual-role factors. Even though models have been developed for working with such ‎dual-role factors, this paper proposes performance appraisal from both an optimistic and ‎pessimistic perspective for selecting a third-party reverse logistics provider in the presence of ‎multiple dual-role factors. A numerical example illustrates the application of the proposed ‎approach.‎

Service organizations such as water and wastewater companies need feedback and performance evaluation to deliver better services. The purpose of this study was to analyze trend of water supply to Iranian cities and villages and wastewater disposal from them during the years 2012 to 2018. For this purpose, in this descriptive-cross-sectional study that required data were adapted from the statistical yearbook of the national water and wastewater company, performance of four areas include urban water supply, rural water supply, urban wastewater disposal and rural wastewater disposal over 7 consecutive years were evaluated by using one of the newest multiple attribute decision making methods called Measurement Alternatives and Ranking according to COmpromise Solution (MARCOS) with considering each year as an alternative and using time series analysis and linear fitting with the help of Excel  and SPSS  software. The results, with a 99% confidence level, showed that the performance of all four areas has significantly increased.