رضا رادفر

Telemedicine, as a growing field, has the potential to fundamentally transform healthcare delivery.

The present study proposes a novel architecture combining IoT data collection systems, decision support systems, and expert physician collaboration to address the challenges in telemedicine and improve the quality of patient identification and treatment.

The proposed architecture enables data integration and simultaneous analysis, and by using Petri nets as a formal modeling tool, it enables system performance and feasibility analysis in complex and real environments and optimizes treatment processes. The proposed model was modeled using CPN Tools software and its performance was evaluated by a panel of experts using a structured Delphi approach, which confirms the results of the feasibility analysis and efficiency of the proposed system. In this regard, with a purposive sampling approach, 12 experts were selected and with the groupthink approach of the expert panel, after identifying six qualitative indicators (data integrity, data quality and security, analysis capability, accuracy and flexibility, general understanding of the model, model innovation, and implementation capability) by designing a closed questionnaire (containing 28 questions), the experts' attitudes were collected and analyzed with the Delphi method approach.

The results of the Delphi method indicate that the data integrity index has the highest weight with an average of 4.01 and the model innovation index has the lowest weight with an average of 3.63.

The expert panel and Delphi methods accept the validation and efficiency of the proposed system from the total evaluation indicators.

The aim of the present study was to develop a paradigmatic model of financial resilience in the water industry using grounded theory methodology. This research was qualitative in nature, and the target population included financial managers, experienced experts, industry leaders in the water sector, policymakers, and university professors. The sample size was selected purposefully and continued until theoretical saturation was achieved from the mentioned statistical population. In this phase, using exploratory semi-structured interviews within the statistical community, the factors and components of financial resilience in the water industry were identified, and with the help of these factors and indicators, a model of financial resilience in the water industry was designed based on grounded theory.According to the results and outputs from the interviews, the factors influencing the financial resilience model in the water industry are as follows:Causal Conditions: Competitive dynamics, improvement of services and products, process optimization, capital increase, differences among client groups, credibility, and bank status.Contextual Conditions: Ownership structure of banks, economic factors, legal and Islamic factors, structural problems of banks.Intervening Conditions: Managerial structure and thinking, alternative investment markets, banking distribution channels, availability of suitable human resources, existing costs, changes in client behavior over time.Strategies: Portfolio of financial products and services in the water industry, financial engineering research and innovation, pricing methods of services, customer relationship management and trust-building based on different client segments, transparency, and performance evaluation of financial services in the water industry.Outcomes: Economic outcomes, macro-level intangible outcomes, tangible industry outcomes, intangible industry outcomes, customer-centered outcomes.

The supply chain is a system consisting of suppliers, manufacturers, distributors, retailers and customers, in which the three main flows of materials, information and financial components interact with each other, in which the financial flow of the supply chain is a fundamental part. It is from the supply chain, which includes all the processes, transactions and transfers related to the flow of cash from the purchase order to the payment to the seller. This research is practical in terms of its purpose, which was done quantitatively. In terms of methodology and the method of data collection, it is considered a descriptive research that has been implemented post-event. In order to collect data in this research, library and document methods were used. The statistical population of this research was Cyrus Food Industries. A meta-heuristic method was used to analyze the data. The results showed that the average of the four criteria of quantity, quality, degree of balance in simultaneously reaching the goals and variety of Pareto solutions obtained from each algorithm were presented. Also, NSGAII algorithm has performed better than NRGA in all four indicators.

an effective mechanism for sharing information with the ability to guarantee privacy and data immutability. At present, permissioned blockchains have made significant inroads into supply chains. In this research, we show that the use of blockchain in the supply chain leads to increased transparency in the supply chain, improved traceability, prevention of disputes, better operational effects and more attention to the environment. In the following, considering the need to improve the housing supply chain in the country, a proposed model for using blockchain in the supply chain of housing construction in the country is proposed based on the existing laws and guidelines, and the implementation steps of its blockchain deployment will be explained.

 Technology development based on strategic alliances in the automotive industry has emerged as a significant strategy in recent decades. This study is descriptive-analytical in terms of data collection method and follows an inductive-deductive approach in terms of research methodology. The qualitative phase involved interviews with 17 experts in the Iranian automotive industry and academic professionals specializing in technology within strategic management. Participants were selected purposefully. The initial interview questions were developed based on a review of books, articles, and related studies. The interviews continued until theoretical saturation was achieved, which occurred at the 15th participant. After transcribing the participants’ responses, an analysis of the explicit and implicit content of the collected information began. The goal of this process was to extract the components required to propose a technology development model based on strategic alliances in the Iranian automotive industry using a grounded theory approach. For analyzing the text of the semi-structured interviews, Strauss and Corbin's (1998) method was employed using MAXQDA software. The findings revealed a total of 203 initial open codes. After removing duplicate codes and merging similar ones, 78 final open codes were identified. Following the analysis of the interview texts, 14 axial codes within six selective codes were extracted. The results indicate that identifying customer needs through data analysis and surveys leads to the design of products aligned with their preferences. Modern technologies such as artificial intelligence and data mining serve as powerful tools for identifying customer behavioral patterns. Aligning product design with customers' cultural and economic needs enhances their loyalty. The experience of successful global companies demonstrates that this approach contributes to increasing market share.

The success of artificial intelligence (AI) deployment in any organization depends on the success of its business intelligence (BI) systems. This research aims to identify the main scenarios for the development of AI-based banking in the country's banks, based on their success in implementing BI systems. The existing research gap arises because current theoretical knowledge cannot explain the failure of AI projects in banks. To formulate these scenarios, a literature review was conducted to extract 32 drivers from the dimensions of BI maturity, whose positive impacts on the successful development of AI applications had been proven in previous studies. The most prioritized of these drivers were then examined by experts from 12 banks that had experience with both BI system deployment and AI-based banking development programs. This was done using a questionnaire and the Fuzzy Delphi technique. In the first stage, 10 drivers were selected, weighted, and normalized using the COPRAS method, and two drivers were chosen from among them: data quality and the level of data and systems integration. Using these two drivers, four future scenarios were formulated. To identify the most probable scenario, the MABAC technique was employed, and experts predicted the likelihood of each scenario based on the BI maturity scores of their respective banks, using the distance from similarity boundary technique. The main finding of the research is that banks are at risk of Costly and Irretrievable Intelligence in the development of AI applications

The shortage of water resources and the increasing pressure on the agricultural sector have made the need to use new technologies to optimize water consumption more apparent than ever before. This research aimed to design a smart irrigation system based on the Internet of Things and a fuzzy case-based reasoning approach, in order to improve the productivity and management of water resources in modern agriculture. The proposed system determines irrigation decisions intelligently and dynamically by using data collected from sensors and processing them by a fuzzy case-based reasoning methodology. The efficiency of this system was compared with two conventional irrigation methods, including surface irrigation and drip irrigation (reactive), based on the criterion of optimization in water consumption, in real conditions of a date palm orchard. Analysis of the results showed that the proposed smart system provides better performance by significantly reducing water consumption and increasing productivity compared to the other two methods. This system also provides the ability to be flexible and adaptable to changing environmental conditions, allowing for precise and optimal management of resources. The use of a smart irrigation system based on the cognitive Internet of Things (IoT) can be used as a sustainable and efficient solution for managing water resources in agriculture.

People's travel behavior is reflected in their choice of transportation modes and is influenced by various factors. Travelling by private car often leads to numerous problems. Therefore, policymaking to shift citizens’ travel behavior from private car use to public bus transportation is important. Hence, the purpose of this research is to examine the travel behavior of Tehran’s residents using a system dynamics simulation model. Accordingly, after identifying the main variables affecting travel behavior through library studies and expert interviews, the hypotheses of the model were formulated. Subsequently, by drawing the cause-and-effect diagram and the stock and flow model, the relevant mathematical equations were derived and validated, and the model was then tested for accuracy and reliability. Subsequently, policies related to the three variables of the number of BRT buses, access to BRT buses and parking capacity were implemented through several scenarios. The results revealed that increasing the rate of parking construction does not lead to favorable results. Moreover, the increase in the number of BRT bus fleet alone cannot have an effective role either under current conditions or when combined with the scenarios involving reducing or increasing parking construction rates. Reducing the rate of parking alone has favorable results. Similarly, increasing the number of BRT stations yields positive results. Moreover, implementing both scenarios simultaneously- reducing the rate of parking development and expanding BRT stations- represents the most effective scenario among those analyzed.

Dental caries is the most prevalent chronic disease worldwide, requiring early diagnosis to prevent invasive treatments. Traditional detection methods have limitations, particularly in detecting early lesions. The aim of this study is to review the advances and challenges of artificial intelligence applications in dental caries detection.

A narrative review of articles published between 2015-2025 was conducted using established scientific databases including PubMed, Scopus, Science Direct, IEEE Xplore, Web of Science, and Google Scholar. Study quality was assessed using the QUADAS-2 tool, and data were analyzed based on artificial intelligence methods, imaging modalities, and performance metrics.

Convolutional Neural Networks (CNNs) and their derivatives such as U-Net, Mask R-CNN, and DenseNet are the most widely used algorithms in caries detection. The diagnostic accuracy of these systems was comparable to or better than dental specialists in many cases, particularly for early lesion detection. Artificial intelligence performance across different dental imaging modalities ranged from 71% to 99.2%.

Artificial intelligence demonstrates significant potential for revolutionizing dental caries detection, particularly in identifying early lesions that are crucial for non-invasive preventive interventions. The optimal approach is "assistive artificial intelligence" that serves as a complement to human clinical judgment rather than a replacement. Main challenges include dataset limitations, lack of standardization in methods and reporting, clinical workflow integration problems, need for transparency and explainability, and ethical-legal concerns. Future directions include developing larger and more diverse datasets, multimodal approaches, improving model explanation methods, conducting longitudinal clinical studies, and developing appropriate standards.

In this article, the requirements and strategies of the circular economy concept in order to create sustainable value in the supply chain of the wind turbine industry have been discussed. In fact, the purpose of the circular economy concept is the highest possible value of a product or a material over time, as well as reducing the consumption of raw materials and increasing the productivity of resources. In this research, the application of the circular economy with the approach of reducing waste, reducing the environmental effects of production and consumption, increasing the productivity and efficiency of resources, and addressing the emerging concerns related to the security and lack of resources in the production of wind turbines, as well as their end of life, are examined. It is expected that wind energy will go through a growing trend in production, installation and commissioning in the world. In the supply chain for wind generators, the highest risk is related to raw materials. Most of the materials used in the production of wind turbines, such are among the critical and strategic materials. In Iran, the capacity of renewable power plants has grown significantly. After wind turbines are decommissioned, materials from wind turbines must be disposed of effectively. In this article, the separation and disposal of these components and materials and the environmental effects and recycling potential of wind farms are evaluated. This paper deals with the relationship between end-of-life, circular economy and life cycle assessment of wind turbines.
 

The purpose of this study is to investigate the systematic effects of marketing mix indicators in distribution resilience. Method of study is descriptive - survey in systematic modeling approach and by considering the relationships among variables, it has been categorized in the mixed - exploratory research group. The samples are two groups, first are experts and second are managers in auto-parts manufacturing. In the qualitative stage, fuzzy Delphi analysis was used for identified indicators and adjust them with the conditions of auto - parts manufacturing by the interview tool, and in the quantitative stage, it was applied by the questionnaire tool. The samples consisted of experts who had the necessary scientific and practical requirements in the supply chain and marketing mix, as well as managers and employees of the country 's automotive auto - parts - manufacturing industry. Snowball sampling and targeted sampling were used in the qualitative and quantitative phases, respectively. System dynamics method and VENSIM software were used to analyze the data, examine the relationships between the variables, and show their effectiveness and flexibility in the dynamics and factor changes in the scenarios. For this purpose, the indicators of marketing how the relationships between them were determined as a model. The results of dynamic analysis showed that marketing factors effect in resilience distribution in supply chain. Four scenarios as changing in innovation and quality, advertising and branding, CRM and marketing expenses, distribution channels and positioning, were designed and tested. Results shown that changing in all indicators have positive effect in resilience, but time and kind of their effects are different.

Today, the use of research findings and their commercialization are considered important priorities. Commercialization of technology is very important for promoting the country's research organizations, promoting economic development and national competitiveness. Therefore, the main goal of this research is to identify the dynamics of the research commercialization ecosystem of agricultural research institutes. In this regard, with library studies and during field research in the form of semi-structured interviews, the dimensions and variables were identified, and then by using the opinions of experts in the field of agricultural research institutes, a dynamic system model was developed and with the help of Vensim software, the model was drawn, formulated, tested and It was simulated, then different scenarios and policies were proposed. According to the results obtained in this research, commercialization desire and ability, knowledge distance, team building strength, entrepreneurial skills and new ideas in the development of agricultural research centers in commercialization were introduced as proposed policies. It is noteworthy that the results of this research can help the development of agricultural businesses and can be used for policy makers and decision makers in the agricultural sector.

This study introduces a Hybrid Markov–Bayesian Framework for predicting and managing accident risks in high-risk industries, with a specific focus on the mining sector. The framework integrates Markov models to analyze dynamic risk transitions and Bayesian networks to infer causal relationships among key human and environmental factors. Drawing from a comprehensive dataset of mining operations, the framework evaluates variables such as age, experience, task type, and injury characteristics to predict and control accident risks. The results highlight the model's high performance, achieving an accuracy of 87%, precision of 85%, and an F1-score of 0.84. This innovative approach enables real-time safety interventions and proactive risk management strategies. The findings underscore the framework's potential to improve workplace safety and serve as a scalable tool for accident prevention in other high-risk industries. Future research will focus on enhancing the framework’s adaptability and incorporating additional contextual variables for broader applicability.

Fumigant gases in maritime and container chains, along with occupational noise in marine and manufacturing industries, are among the most significant chronic risk factors. They are usually assessed separately, despite their simultaneous impact on workers’ health. The importance of this study lies in presenting an integrated approach for real-time monitoring of combined risk and aligning it with occupational exposure limits (OELs). The aim is to develop and validate an interpretable, regulation-oriented framework for predicting combined risk.

This research integrated and normalized data from the Global Burden of Disease (GBD) 2021 study including age-standardized disability rates (ASDR) and average annual percentage change (AAPC) for 204 countries with occupational exposure limit tables for fumigants. A Sugeno-type fuzzy inference system with three inputs and four rules was designed. Weights and membership function boundaries were optimized using the Prairie Dog Optimization algorithm, and a threshold-based scenario generation module was applied to produce high-risk synthetic data. Model performance was evaluated through an OEL compliance test.

Findings revealed that the proposed optimization reduced the loss function by 42% compared to random search. The mean absolute error (0.028 ± 0.006) and root mean square error (0.041) were obtained. Threshold-based scenario generation improved data coverage in high-risk regions from 0.62 to 0.90 and increased the accuracy of critical condition detection from 0.71 to 0.89. The OEL compliance index reached 0.93, confirming input weighting as the most influential factor.

The proposed framework simultaneously ensures numerical accuracy, interpretability, and regulatory compliance with occupational exposure limits. It can be deployed within real-time monitoring dashboards for ports and factories. Future research should integrate IoT sensors and multi-objective optimization to enable dynamic updates in response to evolving regulations and operational conditions.

Designing a smart model for granting banking facilities can reduce many of the concerns and worries of the banking system regarding the repayment of granted loans and the increase in non-performing loans. For this purpose, it is necessary to design such a model using artificial intelligence algorithms. In this research, the aim is to design a smart model for granting banking facilities based on big data and based on demographic characteristics, business status and past payments of customers. The model consisted of 12 variables, the data of which was collected from the banking system and then implemented with the help of machine learning algorithms. Three algorithms, nearest neighbor, decision tree and random forest, were used in the present study, and the results show that the decision tree algorithm is able to predict customer repayment behavior with an accuracy of 99.6% at the highest level, and this indicates that the 12 variables introduced can classify customer repayment based on three categories of customers with high, medium and low risk with more than 99%.

The present article introduces an innovative model (BI.M-LSTM), which combines the BI algorithm and the Long Short-Term Memory (LSTM) neural network to predict flight safety risks during the approach phase. The approach phase, accounting for 3% of the total flight process, is considered the most dangerous stage of any flight. The proposed method involves training supervised neural networks to estimate target parameters. For this purpose, a standardized dataset from the years 2019 to 2020 was used. After summarization, cleaning, and normalization, a total of 28,813 records related to safety risk parameters, such as weather conditions, aircraft configuration, flight information, speed, altitude, and air traffic, were selected. Due to the dependency of flight data on previous inputs and the need for some form of memory, training was performed using the LSTM algorithm in the Python environment. After learning, the mean squared error of deviations was approximately 6.38%. The result showed that the error rate is negligible, and the proposed model has high credibility compared to similar models. This model, equipped with advanced tools including ETL (Extract, Transform, Load), metadata, and real-time monitoring, addressed the challenges of exploring and cleaning large-scale flight data. It successfully predicted the most critical safety factor during the approach phase: speed and altitude control during landing. This reliable approach assists flight crews in controlling important safety parameters, including avoiding loss of control, aircraft speed, touch-down position, and preventing runway excursions.

Analyzing consumer behavior is an effective way to know the characteristics of customers of an economic enterprise and to promote and improve effective marketing strategies. Most retail industries, including chain stores, are connected with a huge collection of unclassified and heterogeneous information, so it is necessary to identify appropriate marketing strategies in order to improve the level of customer loyalty, profit and sales. In this research, the behavior of retail consumers was modeled with a mixed approach (qualitative and quantitative). The statistical population is consumers in the retail industry. In the qualitative part, a conceptual model was designed by the opinions of 22 experts in marketing using an open questionnaire. In the quantitative stage, through simple random sampling and unlimited population with a sample of 112 customers of the store in question, the model designed with the Structural Equation Modeling approach was tested. Based on the results, consumer behavior is a function of causal conditions (product condition, product price, place of purchase, discounts, economic conditions, technological conditions, and cultural conditions), contextual conditions (motivation, advertising, education, and mental image) and intervening conditions (consumer culture, personal characteristics and social relations). These factors provide the conditions for actions and interactions (searching for information and evaluating options) that lead to consequences (product selection, vendor selection, brand selection, purchase amount, purchase time, and payment method). In the quantitative part, information analysis was done in two parts of the measurement and structural model. according to the significant components, the final retail consumer model was presented.

The Impact of Value Added by Small and Medium-Sized Industrial Enterprises (SMEs) on Iran's Economic Growth This study examines the impact of value added by small and medium-sized industrial enterprises (SMEs) on Iran's economic growth. SMEs play a pivotal role in economic development and employment generation due to their ability to create value added, foster innovation, and diversify the economy. Using a structural equation modeling (SEM) approach and a descriptive-applied methodology, this research focuses on a statistical population of 60 small and medium-sized industrial enterprises in Shiraz. The sample size, determined via Cochran’s formula, consisted of 169 senior managers and experts, with 200 questionnaires distributed to ensure adequacy. The questionnaires measured economic value added and economic progress, and their validity and reliability were confirmed. The findings reveal that SMEs significantly improve the country’s economic indicators. Notably, research activities and technological innovations exhibit the highest factor loading (0.262), playing a vital role in economic growth and advancement. Additionally, employment (factor loading: 0.225) and income distribution (factor loading: 0.243) contribute to improved labor conditions and equitable income distribution. Investments in these enterprises (factor loading: 0.148) further support sustainable economic growth. However, despite the importance of exports in economic development, this variable was excluded from the study. Thus, future research should investigate the impact of SME exports on Iran’s economy to provide a more comprehensive analysis of their role in economic development.

This research is qualitative in terms of research method; In terms of research approach, inductive; In terms of the research paradigm, pragmatism; In terms of research strategy, data theory is the foundation, and in terms of data collection source, semi-structured interview. Using the purposeful sampling method, 15 people were selected as a statistical sample; this number was considered based on the saturation law. In the quantitative part of the statistical community, 120 managers of service start-ups based on information technology were selected in 1402, and the final consolidated model was obtained with the method of interpretive structural model. The purpose of this article is to formulate a strategy model for the success of Iran's startups businesses in the field of information technology. The results of the data analysis showed that the three main dimensions as the main dimensions of the failure of startup businesses in Iran are: factors of the external business environment, factors of the internal business environment and factors related to the founders, management team and human resources. Legal and international restrictions, macro-government policies, and structural and infrastructure factors are among the most important factors in the failure of startup businesses in Iran. Also, the government's support increases financial strength and improves liquidity in startup businesses. Appropriate upstream policies and planning in the way of facilitating the business environment can help the success of startup businesses.

Development of artificial intelligence in broad industrial and commercial fields, and significant increasing of non-linear data which produced by various units of organizations, on the other hand, would be necessary for using artificial intelligence (IA) in inventory management (IM).That's why the strategic model recommended in this article aims to use deep learning methods of long-term memory (RNN) and bidirectional long-term short-term memory (BILSTM) neural network based on regular inventory control methods such as the economic value of the order( EOQ) and the economic production quantity model (EPQ) also use of ABC analysis which facilities would increase effectiveness of model. The achievement of this article is reach a unique strategic model, by integration of traditional inventory control methods and artificial intelligence, which has high efficiency in terms of speed and accuracy of forecasting. However the findings have been showed that the recurrent neural network(RNN) ​​with an accuracy of 85% did not acceptable and combined approach of the (BILSTM) neural network, after building and training by considered parameters in line with the purpose of the article, has explained the accuracy of 93%.

This systematic literature review aims to explore and analyze the application of the system dynamics approach in the LARG-Effective Supply Chain Model. The LARG-Effective Supply Chain Model is a comprehensive framework that integrates various factors affecting supply chain performance. This paper overviews lean, agile, resilient, and green supply chain management principles. Lean principles focus on eliminating waste and improving efficiency in processes, while agile principles emphasize flexibility and responsiveness to changing market demands. Resilience principles aim to build robustness and adaptability in supply chains to withstand disruptions. Green supply chain management focuses on reducing environmental impact through sustainable practices. Also, this research explores how these principles can be integrated into a cohesive framework for managing sustainable supply chains. The authors discuss various strategies for integrating lean, agile, resilience, and green practices, such as using lean tools to identify waste in processes and implementing agile strategies to respond quickly to changes in customer demand. Moreover, the paper examines the role of the system dynamics approach in modeling and analyzing complex interactions within sustainable supply chains. System dynamics is a powerful tool for understanding how different variables within a system interact with each other over time. By using system dynamics models, researchers can simulate various scenarios and test different strategies for improving sustainability in supply chains. By examining relevant studies, this review seeks to identify the benefits, limitations, and potential areas for improvement in utilizing system dynamics within this model. The findings of this review will contribute to a better understanding of how system dynamics can enhance supply chain management practices.

Inventory management as a live subset of whole organization would be organize the expectation of supply chain base on an intelligent platform. This article was created with the aim of facing the challenges of industrial big data on data which are effective in decision-making based on historical records, but are not linear data due to the valuable content they have.As consequence, it is obvious that are not exploited by traditional methods such as time series for forecasting and decision-making. Accordingly, in this article, to present a strategic model based on artificial intelligence techniques, long short-Term Memory (LSTM), while we discussed the use of industrial big data, which enables the ability to use non-linear data to predict orders. Moreover, the methods of economic order quantity (EOQ) and economic production quantity model (EPQ) in order to prepare and training data have been used. In continue for getting better processing in order to identify strengths and weaknesses in inventory management, "ABC" analysis has also been used. The findings show that the integration of the traditional approach with deep learning method improves learning performance for complex big data analysis tasks. Many evaluations of the proposed deep network have reached an acceptable error value. Finally, this model has provided promising results for predicting expectation of fields that related to controlling amount of production based on objective in inventory management sector.

Dental caries is one of the most prevalent chronic oral diseases worldwide. Timely and accurate diagnosis of dental caries plays a crucial role in preventing lesion progression and reducing complications. This study aimed to systematically review the studies on dental caries detection using machine learning algorithms applied to periapical radiographs.

A comprehensive search was conducted in PubMed, Scopus, Web of Science, IEEE Xplore, and Google Scholar databases up to the end of 2024. Inclusion criteria comprised studies using machine learning algorithms for detecting dental caries in periapical or intraoral radiographs. The quality of studies was assessed using the QUADAS-2 tool.

From 825 initial articles, 13 studies met the inclusion criteria. All studies used Convolutional Neural Networks (CNNs) with various architectures including ResNet, VGG, Inception, DenseNet, and YOLO. ResNet-based models and their hybrid variants showed the best performance with diagnostic accuracy ranging from 82% to 98%. Comparison with human experts in 6 studies revealed that deep learning algorithms demonstrated similar or superior performance.

From the results, deep learning especially convolutional neural networks, had significant potential for improving dental caries detection in periapical radiographs. However, challenges such as limited high-quality training data and generalizability issues need further investigation.