نشریه مهندسی مدیریت نوین
سال دهم شماره 3 (پاییز 1403)

Deep learning is a subset of the broader class of neural network based machine learning methods that has recently gained much attention in many fields, including time series prediction in financial markets. In this research, first, using deep learning models based on LSTM and CNN networks, the movement of Tehran Stock Exchange index is predicted. Then, by combining the two models, the hybrid CNN-LSTM deep learning model is used to predict the Tehran Stock Exchange index. In the next step, in order to evaluate the performance of the aforementioned forecasting models, three performance measures symmetric mean absolute error percentage (SMAPE), mean absolute error percentage (MAPE) and root mean square error (RMSE) were used. In this research, the daily data of the Tehran Stock Exchange Index was used in the period of 2016-07-13 until 2021-01-26. The estimation results of the models in predicting the Tehran Stock Exchange index with a one day step and comparing the efficiency measurement criteria indicate the superiority of the proposed CNN-LSTM model compared to the other two models. The LSTM model ranks next in accuracy and forecasting efficiency. According to the results that be presented in this research, financial market participants in Iran are suggested to pay attention to integrated deep learning models in order to increase the efficiency and accuracy of their predictions.

In this research, a model of optimal control for customer dynamics based on marketing policies is investigated as a non-automatic system of differential equations. The main purpose of the model is to track and analyze the simultaneous changes in the behavior of regular, referral and potential customers of the company from the time of its inception to now. Implementing an effective marketing policy to optimize these changes and increase the number of customers is of particular importance. In line with this goal, a new supervised machine learning algorithm is presented for the numerical simulation of the problem. The proposed algorithm uses polynomial kernels. Polynomial kernels make it possible to simulate a complex function of data in a way that helps to better understand customer dynamics. Support Vector Least Squares regression provides a simple optimization method for marketing strategies, with this approach, marketing strategies can be optimized without dealing with the details of each customer and instead focusing on the overall effect of this strategy. placed on the set of customers. This research shows how machine learning techniques can help in solving complex management and marketing problems. Over time, the number of regular customers increases and the number of potential customers decreases. However, the number of referral customers shows a rapid growth at the beginning of the time period and a fluctuating increasing pattern over time.

This research aimed to explore organizational socialization in the ecosystem of public education in Iran. This study was conducted using a mixed-method (qualitative-quantitative) approach. The qualitative part employed content analysis with an interpretive approach, following Attride-Stirling’s approach. The research environment in this section included articles and books in the field of organizational socialization. Data analysis in the qualitative section revealed the dimensions of organizational Socialization, including organizational adaptability, active behavior, organizational culture, future envisioning, and organizational proceduralism. The quantitative part used a descriptive survey method. The statistical population in the quantitative section consisted of teachers employed in different grades of public schools in selected provincial centers in 2021, who were selected through convenience sampling. The research instrument was a 127-item questionnaire derived from the qualitative part of the research and based on a 5-point Likert scale with content validity confirmed by subject matter experts and face validity established by several members of the target population. Structural validity was assessed through confirmatory factor analysis, and reliability was measured with Cronbach's alpha, which exceeded 0.70 for all items. Quantitative data were analyzed using descriptive and inferential statistics, including factor analysis. Based on the factor loadings obtained for each item, the research model demonstrated desirable validity. To promote organizational socialization for novice and experienced teachers, it is recommended to utilize the actions derived from this research in organizational policy levels and teacher training programs.

The aim of the present study is to identify the factors that affect internal branding with emphasis on the cultural approach of the Islamic Azad University, Khorasgan Branch. The research method is qualitative research (ground theory) and is applied in terms of purpose. Interviews were used to collect information. The external validity and qualitative reliability of the data have been calculated and confirmed. The statistical population was a group of experts and professors of Khorasgan University, which was initially 9 people and then continued until the final saturation of 12 people. The ground theory method was used to analyze the data. First, categories and concepts were extracted through interviews and with the help of open coding and central coding, and the number of 91 open codes and 29 central categories was obtained. Internal branding is to increase employees' commitment to the organization; Of course, this will be achieved if the senior managers support the activities of human resources management and also the employees of the organization. The more this support, the commitment of the employees in the organization will increase and the internal brand will be better formed. Also, human resource managers were familiar with all factors of human resource management, such as the processes of recruitment, retention, performance evaluation, training, development, health and safety, communication, suggestion system, etc., and their main problem is the lack of implementation of these processes due to various reasons such as culture.

The purpose of this research is to investigate the effect of using the poster advertising model in the field of public health. In terms of purpose, this research is descriptive-survey in the category of applied research and based on the method of data collection. The statistical population of this research includes all managers of health centers in Isfahan province who were employed in 1401 and their number was 249 people, of which 150 people were selected as the research sample by stratified random sampling method. A researcher-made questionnaire was used whose validity and reliability were confirmed. To analyze the collected data, Kolmogorov-Smirnov tests were used for the normality of the data, as well as confirmatory factor analysis and structural equation method using the software (AMOS24).The findings showed that among the studied components, respectively: solutions with path coefficient (0.80), effectiveness (0.84), ruling platforms (0.92), obstacles with path coefficient (0.94) and effective components (0.97) are most effective in the poster advertising model in the field of public health. The model obtained from this research can open the way for practitioners and planners of the health and treatment sector to plan for the effectiveness of poster advertising campaigns in the field of public health and to improve the level of public health and health, as well as to create innovative opportunities to increase productivity in this field.

Solid waste has become one of the most critical environmental issues in the world. Therefore, a waste management system to prevent further destruction of the environment is essential. Waste management includes collection, transport, cleaning, recycling, and disposal of the wastes. In recent years, due to environmental concerns, manufacturers have been forced to offer environmentally friendly products. So, the area of reverse logistics (RL) has recently received considerable attention, due to a combination of environmental, economic, and social factors. In this research, the design of a multi-product and eleven-level reverse logistics network is conducted, which collects all the waste in one place and separates them according to the needs of the factories (in terms of the type and material of the waste, etc.) and sends them to the intended destination. This model can support all kinds of industries in which the revival of recycling and destruction of products. This study provides a mixed integer mathematical model to reduce the costs of the whole system. The number of centers, the number of products and parts that should be sent from one center to another, the amount of CO2 emissions, and the total cost of the model were determined. Finally, the sensitivity analysis was done on the parameters of the model. The model was validated by changing the input data in two different cases.