درخت حوزه‌های تخصصی

Purpose – The purpose of this paper is to optimize Cost & time in the three echelon supply chain (SC) network. This paper developed a linear programing (LP) model to consider economic data. Design/methodology/approach – The overall objective of the present study is to use high-quality raw materials, at the same time in post COVID 19 pandemic and the highest profitability is achieved. The model in the problem is solved using two metaheuristic algorithms, namely, Cuckoo and Genetic. Optimization of supply chain performance indicators in minimization of cost and time and maximization of sustainability indexes of the system. Findings – The differences found between the genetic algorithms (GAs) and the LP approaches can be explained by handling the constraints and their various logics. To deal with ambiguity in the reverse logistics network, a fuzzy approach has been applied. To solve the problem in large dimensions, meta-heuristic algorithms of Cuckoo and Genetic were employed by applying MATLAB software. In order to compare two optimization algorithms, a series of sample problems have been generated then the results of two algorithms were compared and superiority of each of them was discussed.

This article deals with developing a model of digital marketing with a focus on customer experience and with the aim of optimizing the customer management process, increasing the quality and length of the customer journey (increasing loyalty), and improving the consequences of the marketing process in companies, and seeks to answer this question. What is the digital marketing model with the customer experience management approach and what components does it include? The research method is mixed (qualitative-quantitative) and since a model for this issue is not available in past research, the first part of the research explores model extraction through qualitative methods (here, foundation data analysis) and to determine Relationships within the model quantitatively explain the model using quantitative methods (here structural equations). The method of collecting information in the qualitative part, in addition to conducting library studies, in the form of in-depth interviews with experts in the fields of information technology, marketing, and commerce, information technology policy, and actually business managers and stores that provide services and products through channels and digital tools in this section, the statistical sample limit was carried out by the snowball method (saturation to the end of the problem) and the quantitative part of the research and explanation of the model was through data collection with questionnaire tools and closed interviews from the customers of several digital businesses, which was done with the method 385 people were randomly selected and available using Cochran's formula. In the end, the main conditions, background conditions, intervening conditions, phenomena, strategies, and consequences affecting the digital marketing model with the customer experience management approach were extracted and the relationships between them were determined.

The purpose of the present study is to use the ensemble learning model to combine the predictions of Random Forest (RF), Long-Short Term Memory (LSTM), and Recurrent Neural Network (RNN) models for the Top and Bottom Prices of Bitcoin. To this aim, in the first stage, Bitcoin's top and bottom prices are predicted using three machine learning models. In the second stage, the outputs of the models are presented as feature variables to the Extreme Gradient Boosting (Xgboost) and Light Gradient Boosting Machine (LightGBM) models to predict the price tops and bottoms. Then, in the third stage, the outputs of the second stage are combined through the voting ensemble classifier pattern to predict the next top and bottom prices. The data of top and bottom Bitcoin prices in the 1-hour time frame from 1/1/2018 to the end of 6/30/2022 are used as target variables and 31 technical analysis indicators as feature variables for the three models in the first stage. 70% of the data is regarded as learning data, and the remaining 30% is considered for the second and third stages. In the second phase, 50% of the data is considered for learning the output of the previous stage and 50% for the test data. Finally, the prediction values are evaluated with real data for the three models and the proposed ensemble learning model. The results reveal the improvement of the performance, precision, and accuracy of the ensemble model compared to weak learning models.