Data-driven Decision Research Articles

IoT and Edge Computing for Smart Manufacturing: Architecture and Future Trends

The integration of the Internet of Things (IoT) and Edge Computing is revolutionizing the manufacturing industry, ushering in the era of smart manufacturing as part of Industry 4.0. This paper explores the synergy between IoT and Edge Computing, focusing on their combined architecture and the future trends driving innovation in smart factories. IoT enables the connection and communication of machines, sensors, and systems, allowing for real-time data collection and monitoring. However, traditional cloud-based approaches face challenges such as latency, bandwidth limitations, and security risks, which can hinder real-time decision-making in fast-paced manufacturing environments. Edge Computing addresses these issues by processing data closer to the source, minimizing latency and reducing dependence on cloud infrastructures. By combining IoT and edge solutions, smart manufacturing systems can make faster, data-driven decisions, leading to improved efficiency, reliability, and operational flexibility. This paper delves into the architectural design of IoT and edge computing in manufacturing, outlining how data flows from IoT devices to edge nodes and cloud services. Several real-world use cases and industry examples are analyzed to highlight the practical benefits of these technologies. Additionally, this research identifies key challenges such as security vulnerabilities, the need for robust network infrastructures (e.g., 5G), and issues related to data standardization. The future of smart manufacturing is also explored, emphasizing trends like the adoption of artificial intelligence (AI) and machine learning (ML) at the edge, digital twins for real-time monitoring, and the role of IoT and edge computing in fostering sustainability through energy-efficient production processes.This study provides a comprehensive overview of IoT and edge computing architectures in smart manufacturing and offers insights into future technological trends that will shape the industry.

Effectiveness of Generative AI for Post-Earthquake Damage Assessment

After an earthquake, rapid assessment of building damage is crucial for emergency response, reconstruction planning, and public safety. This study evaluates the performance of various Generative Artificial Intelligence (GAI) models in analyzing post-earthquake images to classify structural damage according to the EMS-98 scale, ranging from minor damage to total destruction. Correct classification rates for masonry buildings varied from 28.6% to 64.3%, with mean damage grade errors between 0.50 and 0.79, while for reinforced concrete buildings, rates ranged from 37.5% to 75.0%, with errors between 0.50 and 0.88. Fine-tuning these models could substantially improve accuracy. The practical implications are significant: integrating accurate GAI models into disaster response protocols can drastically reduce the time and resources required for damage assessment compared to traditional methods. This acceleration enables emergency services to make faster, data-driven decisions, optimize resource allocation, and potentially save lives. Furthermore, the widespread adoption of GAI models can enhance resilience planning by providing valuable data for future infrastructure improvements. The results of this work demonstrate the promise of GAI models for rapid, automated, and precise damage evaluation, underscoring their potential as invaluable tools for engineers, policymakers, and emergency responders in post-earthquake scenarios.

Operational Excellence: Process optimization for competitive edge

This study explores the concept of operational excellence and the role of process optimization in gaining a competitive edge for organizations. Operational excellence involves the continuous improvement of business processes to increase efficiency, reduce waste, enhance quality, and improve overall performance. The study reviews established methodologies such as Lean, Six Sigma, and Total Quality Management (TQM) that provide structured approaches to achieving operational efficiency. It also examines the impact of emerging technologies, including artificial intelligence (AI), machine learning (ML), and the Internet of Things (IoT), in driving data-driven decision-making and enhancing process optimization. Additionally, the study highlights the importance of leadership in fostering a culture of continuous improvement and collaboration across departments, ensuring alignment with strategic objectives. The integration of sustainability into operational strategies is discussed as a means to not only improve efficiency but also meet corporate social responsibility goals. The relationship between process optimization and innovation is also explored, showing how operational excellence can free up resources for research and development (R&D) and foster a culture of innovation. Despite the focus on these established practices and technologies, the study acknowledges limitations, particularly the challenges faced by small and medium-sized enterprises (SMEs) and service oriented sectors in implementing these strategies. The study concludes that operational excellence, supported by technological advancements and a strong organizational culture, is essential for organizations to navigate competitive markets, improve customer satisfaction, and achieve long-term success. This research offers valuable insights for businesses aiming to enhance operational performance and sustain their competitive advantage through continuous process optimization and innovation.

Analisis Pengambilan Keputusan JAKI terhadap Peningkatan Pelayanan Publik di Provinsi DKI Jakarta

This research aims to explore the implementation of the intelligence phase in the development of the Jakarta Kini (JAKI) application as an integral part of the Jakarta Smart City initiative. The intelligence phase, described by Herbert Simon as a crucial stage in the decision-making process, involves identifying the problem, collecting relevant data, and deeply understanding the context. Using a descriptive qualitative method approach, this research aims to dig deeper into how Jakarta Smart City implements the intelligence phase to improve electronic-based public services through the JAKI platform. The results show that the effective implementation of the intelligence phase has contributed significantly to improved quality of public services provided, better responsiveness to community needs, and improved data-driven decision-making in the development of JAKI. The implications of these findings are highly relevant and provide valuable insights for other cities seeking to develop similar smart city initiatives. By integrating the intelligence phase in the decision-making process, it is expected to improve efficiency, service quality, and responsiveness to community needs in an evolving and changing urban context. As such, this research makes a substantial contribution to the understanding of the implementation of the intelligence phase in the context of the development of the Jakarta Kini (JAKI) application as part of the Jakarta Smart City initiative.

Machine learning-aided hybrid technique for dynamics of rail transit stations classification: a case study

Accurate classification of rail transit stations is crucial for successful Transit-Oriented Development (TOD) and sustainable urban growth. This paper introduces a novel classification model integrating traditional methodologies with advanced machine learning algorithms. By employing mathematical models, clustering methods, and neural network techniques, the model enhances the precision of station classification, allowing for a refined evaluation of station attributes. A comprehensive case study on the Chengdu rail transit network validates the model’s efficacy, highlighting its value in optimizing TOD strategies and guiding decision-making processes for urban planners and policymakers. The study employs several regression models trained on existing data to generate accurate ridership forecasts, and data clustering using mathematical algorithms reveals distinct categories of stations. Evaluation metrics confirm the rationality and accuracy of the results. Additionally, a neural network achieving high accuracy on labeled data enhances the model’s predictive capabilities for unlabeled instances. The research demonstrates high accuracy, with the Mean Squared Error (MSE) for regression models (Multiple Linear Regression (MLR), Deep-Learning Neural Network (DNN), and K-Nearest Neighbor (KNN)) remaining below 0.012, while the neural networks used for station classification achieve 100% accuracy across seven time intervals and 98.15% accuracy for the eighth, ensuring reliable ridership forecasts and classification outcomes. Accuracy in rail transit station classification is critical, as it not only strengthens the model’s predictive capabilities but also ensures more reliable data-driven decisions for transit planning and development, allowing for more precise ridership forecasts and evidence-based strategies for optimizing TOD. This classification model provides stakeholders with valuable insights into the dynamics and features of rail transit stations, supporting sustainable urban development planning.

Internal Quality Control and the Role of Learning Management Systems (LMS) in Higher Education Institutions in Sierra Leone: Then, Now, and Later

The paper explores the evolution of Internal Quality Control (IQC) in Sierra Leone's higher education institutions (HEIs) with a specific focus on the transformative role of Learning Management Systems (LMS). Beginning with the colonial era, the paper traces the historical context of IQC, emphasizing its informal nature. The post-independence expansion of HEIs introduced quality challenges compounded by economic constraints and political instability. The civil war further disrupted the education sector, and the reconstruction period saw the establishment of the Tertiary Education Commission (TEC), formalizing IQC processes. The introduction of LMS platforms, such as Moodle and Google Classroom, marked a pivotal shift in enhancing educational quality. LMS has significantly contributed to data-driven decision-making, standardized educational processes, and continuous feedback mechanisms. However, challenges remain, including infrastructural limitations and the urban-rural digital divide. The paper also projects the future of IQC, highlighting the potential integration of advanced technologies such as artificial intelligence, big data analytics, and blockchain in LMS platforms. These advancements promise to revolutionize the educational experience, offering new possibilities for quality assurance and student engagement.

Research on Data Quality Analysis Based on Data Mining

As the process of digitalization continues to advance, big data plays a pivotal role across various industries. However, the quality of data directly impacts the results of data analysis and the accuracy of decision-making. High-quality data can provide a reliable basis for decisions, whereas poor-quality data may lead to erroneous judgments and cause incalculable losses. Therefore, the management and enhancement of data quality have become central issues in contemporary data management. Data mining technology, as a powerful tool, excels not only in extracting valuable information from large-scale datasets but also in demonstrating significant potential in improving data quality. By mining and analyzing the patterns and characteristics hidden within data, it is possible to identify and rectify defects, thereby enhancing the overall quality of data and increasing the reliability of data-driven decision-making. This paper delves into the definition and dimensions of data quality, explores the fundamental principles of data mining technology, and examines its specific applications in data quality analysis, aiming to offer new insights and methods for data quality management.

Spatial-temporal deep learning method for solving data-driven multi-echelon stochastic lot sizing problems with intermediate demands

ABSTRACT For the production processes and inventory decisions of a multichannel supply chain, a multi-echelon stochastic lot sizing problem with intermediate demands is extended to a constrained problem that considers the production (inventory) capacity constraint and to multi-item problems that consider feature and item correlations. Three spatial-temporal deep learning methods combining convolutional, recurrent and graph convolutional neural networks are proposed to transform the extended problems as spatial-temporal sequential prediction problems and learn data-driven decisions. Real-world data are used in the computational experiments, and the results show that considering both temporal and spatial features and/or feature correlations can provide improved prediction performance.

Leveraging SAP Analytics Cloud for Data-Driven Decision Making in Modern Enterprises

In today's fast-paced business environment, organizations are increasingly relying on data analytics to make informed decisions. SAP Analytics Cloud (SAC) has emerged as a leading solution for integrating data from various sources, visualizing trends, and driving data-driven decisions. This paper examines the functionalities of SAC, its impact on decision-making processes, and how enterprises use it to gain a competitive edge. Key features such as predictive analytics, machine learning, and real-time data processing are discussed. The paper also includes case studies that illustrate the practical application of SAC in enterprise environments.

Optimizing emergency department efficiency: a comparative analysis of process mining and simulation models to mitigate overcrowding and waiting times

ObjectiveOvercrowding and extended waiting times in emergency departments are a pervasive issue, leading to patient dissatisfaction. This study aims to compare the efficacy of two process mining and simulation models in identifying bottlenecks and optimizing patient flow in the emergency department of Al-Zahra Hospital in Isfahan. The ultimate goal is to reduce patient waiting times and alleviate population density, ultimately enhancing the overall patient experience.MethodsThis study employed a descriptive, applied, cross-sectional, and retrospective design. The study population consisted of 39,264 individuals referred to Al-Zahra Hospital, with a sample size of at least 1,275 participants, selected using systematic random sampling at a confidence level of 99%. Data were collected through a questionnaire and the Hospital Information System (HIS). Statistical analysis was conducted using Excel software, with a focus on time-averaged data. Two methods of simulation and process mining were utilized to analyze the data. First, the model was run 1000 times using ARENA software, with simulation techniques. In the second step, the emergency process model was discovered using process mining techniques through Access software, and statistical analysis was performed on the event log. The relationships between the data were identified, and the discovered model was analyzed using the Fuzzy Miner algorithm and Disco tool. Finally, the results of the two models were compared, and proposed scenarios to reduce patient waiting times were examined using simulation techniques.ResultsThe analysis of the current emergency process at Al-Zahra Hospital revealed that the major bottlenecks in the process are related to waiting times, inefficient implementation of doctor’s orders, delays in recording patient test results, and congestion at the discharge station. Notably, the process mining exercise corroborated the findings from the simulation, providing a comprehensive understanding of the inefficiencies in the emergency process. Next, 34 potential solutions were proposed to reduce waiting times and alleviate these bottlenecks. These solutions were simulated using Arena software, allowing for a comprehensive evaluation of their effectiveness. The results were then compared to identify the most promising strategies for improving the emergency process.ConclusionIn conclusion, the results of this research demonstrate the effectiveness of using simulation techniques and process mining in making informed, data-driven decisions that align with available resources and conditions. By leveraging these tools, unnecessary waste and additional expenses can be significantly reduced. The comparative analysis of the 34 proposed scenarios revealed that two solutions stood out as the most effective in improving the emergency process. Scenario 19, which involves dedicating two personnel to jointly referring patients to the ward, and scenario 34, which creates a dedicated discharge hall, have the potential to create a more favorable situation.

Artificial Intelligence and Machine Learning with Moment Generating Functions to Enhance Biological Count Data Analysis

This research presents artificial intelligence with supervised machine learning to determine the median count given training data from biological laboratory experimentation. After determining the central moment of the median, machine learning is then used to predict the population median value. When computer programs learn they access available data, then autonomously analyze the information to make informed, data-driven decisions. Moment generating function restrictions identify a parameter of interest by restricting the expected value of the moment generating function. This research proposes a theoretical foundation for achieving such predictions. A summary of the key elements underlying the statistical power of the Mann– Whitney test, a nonparametric hypothesis test for the median of count data, is also presented. A nonparametric approach allows for accurate predictions while relaxing distributional assumptions such as normality.

AI-Enhanced GraphoText Analysis for Tracking Counselling Therapy Progress: Integrating Multimodal Graphology and Machine Learning

Despite the robustness of existing instruments, routine manual observation in psychotherapy faces significant challenges, including lack of time, labour, high implementation costs, and potential biases such as response bias and social desirability. This study proposes a dynamic assessment approach using machine learning and multimodal features extracted from handwriting, incorporating graphology-based and content-based features. The integration of content-based and graphology-based features involves combining text and handwriting features with the Support Vector Machine (SVM) using the Radial Basis Function Kernel (RBF). The results show that this approach achieves an impressive accuracy rate of 86.25%. The proposed framework not only improves psychotherapeutic practise, but also offers new insights into human cognition and emotional dynamics by revealing intricate patterns in handwriting. This advance facilitates data-driven decision-making, improves the quality of patient care, and overcomes challenges associated with manual monitoring, including social desirability bias and response set bias. This research paves the way for innovative methods at the interface of mental health and technology, and promises a more objective and efficient approach to monitoring the progress of psychotherapy.

Enhancing the efficiency of management of transport enterprises: modern trends and challenges

This research examines the modern trends and challenges in enhancing the efficiency of management in transport enterprises, with a focus on the Ukrainian perspective and the impact of the ongoing war. The study highlights key trends such as digital transformation, sustainability efforts, and the increasing use of data-driven decision-making, which are reshaping the transport industry in Ukraine. The integration of technologies like AI-based route optimization and real-time tracking has become important in overcoming operational challenges caused by the war, including disrupted supply chains, damaged infrastructure, and heightened security risks. These innovations are enabling transport enterprises to minimize costs, optimize routes, and ensure service continuity despite the unstable environment. This paper investigates how Ukrainian transport enterprises are adapting to these modern trends while addressing the complex challenges posed by the war. It emphasizes the need for a flexible and resilient management approach that leverages digital tools, sustainability practices, and data analytics to enhance operational efficiency. The findings highlight that a systemic approach, supported by innovation and adaptability, is essential for overcoming the impact of war and ensuring the long-term viability of transport sector of Ukraine.

Implementation of Artificial Intelligence in Quality Management in SMEs: Benefits and Challenges

The adoption of artificial intelligence (AI) in small and medium-sized enterprises (SMBs) has emerged as a key strategy for improving quality management. This article discusses the main benefits and challenges of implementing AI in this context. A systematic review of the academic literature and relevant business reports was carried out, in order to collect and analyze empirical evidence on the application of AI in the quality management of SMEs, obtaining as results that the main benefits of the implementation of AI in the quality management of SMEs include: Improved data-driven decision-making, process automation and error reduction, early detection of quality issues, customization of products and services, and resource optimization and cost reduction. On the other hand, the main challenges were: Lack of knowledge and skills in AI, high implementation costs, concerns about data security and privacy, resistance to cultural change, and integration of AI with existing systems and processes, concluding that the implementation of AI in SMB quality management offers significant benefits, but it also entails challenges that must be addressed strategically. SMEs that manage to overcome these challenges will be able to take advantage of the competitive advantages offered by AI, improving their efficiency, quality and adaptability in an increasingly dynamic business environment.

Efficient prediction uncertainty quantification in dam behavior monitoring with attention-based sequence-to-sequence learning

Displacement is an intuitive monitoring indicator of dam structural behavior. Conventional deterministic modeling methods frequently disregard the inherent uncertainty associated with monitoring data. This study aims to quantify such uncertainties within dam displacement predictions by augmenting a sequence-to-sequence structure with the novel recurrent unit, termed as tiny gated unit (TGU), attention mechanism, as well as quantile regression, resulting in the Att-S2STQ model. Specifically, TGU adopts only one gate to identify nonlinearity in data sequences, while Bahdanau attention mechanism dynamically assigns weights to different parts of the input sequence. The inclusion of quantile regression allows Att-S2STQ to produce prediction intervals (PIs), from which the probability density functions (PDFs) and cumulative distribution functions (CDFs) are further derived via kernel density estimation. PIs, PDFs, and CDFs jointly serve to reveal the prediction uncertainty. The effectiveness of the model is illustrated through comparative experiments using real-world dam monitoring datasets. Results indicate that the proposed model outperforms traditional methods in point, interval, and probability predictions, while also having a simpler structure and faster training. The superiority of both accuracy and efficiency makes it a valuable tool for dam management, aiding in data-driven decision-making and enhancing operational safety.

Development of an integrated model for SME marketing and CRM optimization

The development of an integrated model for optimizing marketing and customer relationship management (CRM) in small and medium-sized enterprises (SMEs) presents a crucial advancement in enhancing business competitiveness and customer engagement. This paper proposes a unified model that merges scalable software solutions for social media analytics with AI-driven CRM to provide a comprehensive framework for SMEs to enhance their marketing strategies and customer interactions. Social media platforms have become vital channels for customer outreach, engagement, and feedback, yet SMEs often face challenges in leveraging these platforms due to limited resources and expertise. Integrating social media analytics into CRM systems offers businesses the ability to analyze customer behaviors, preferences, and sentiment on a large scale. The proposed model utilizes artificial intelligence (AI) to streamline customer interactions, enabling personalized communication and improving customer retention. By incorporating AI into CRM, SMEs can automate routine tasks, such as responding to customer inquiries and segmenting audiences based on their social media activity. This allows for more efficient marketing strategies and enables businesses to proactively address customer needs, ultimately driving customer loyalty and long-term growth. Furthermore, the unified model focuses on scalability, ensuring that SMEs can adopt and adapt these technologies according to their evolving needs and resources. This paper also explores the benefits of integrating these technologies, including the potential for improved data-driven decision-making, enhanced customer experience, and increased operational efficiency. In conclusion, this integrated model presents a strategic pathway for SMEs to optimize their marketing and CRM efforts through the adoption of scalable, AI-driven tools that enhance customer engagement and boost business performance. Keywords: SME Marketing, CRM Optimization, Social Media Analytics, AI-Driven CRM, Customer Engagement, Scalable Software Solutions, Business Performance, Personalized Communication, Customer Retention, Data-Driven Decision-Making.

The Role of Edge Computing in Driving Real-time Personalized Marketing: a Data-driven Business Perspective

The fast pace of marketing innovation in the digital age has sparked a growing need for instant, personalized conversations with customers. This study investigates the vital significance of edge computing in elevating real-time personalized marketing by reducing latency, improving speed of data processing, and allowing companies to quickly, data-driven decisions. Through a rigorous investigation, the research reports on the ways edge computing solves problems linked to cloud solutions, like bandwidth issues and privacy matters, providing a more effective framework for localized data management. Conducting a mixed-methods study using case studies and quantitative performance metrics, this research showcases important enhancements in consumer engagement and conversion rates amongst enterprises that have implemented edge computing for their personalized marketing efforts. Findings indicate a reduction of 30% in data transfer latency and a 25% higher level of interactions with customers in real time as a result of marketing activities made possible by the edge. The paper wraps up with a look at the innovation of this method in the face of technological innovations and presents recommendations for firms attempting to raise consumer engagement via personalized marketing approaches facilitated by edge computing.

6778 Visual Presentations for Adherence Risk Management In Patients Receiving Recombinant Human Growth Hormone Therapy

Abstract Disclosure: P. van Dommelen: Consulting Fee; Self; PvD has a consultancy agreement with Merck KGaA, Darmstadt, Germany. F. Michelis: Employee; Self; FL is an employee of Ares Trading SA, an affiliate of Merck KGaA. L. Arnaud: Employee; Self; LA is an employee of Ares Trading SA, an affiliate of Merck KGaA. S. Loche: Advisory Board Member; Self; SL has received advisory board fees from Merck KGaA, Darmstadt, Germany. Consulting Fee; Self; SL has received consultancy fees from Merck KGaA, Darmstadt, Germany. Speaker; Self; SL has received lecture fees from Merck KGaA, Darmstadt, Germany. E. Koledova: Employee; Self; EK is an employee of Merck KGaA,Darmstadt, Germany. Stock Owner; Self; EK holds shares in Merck KGaA, Darmstadt, Germany. Background: A data-driven clinical decision support system based on visual presentations for adherence risk management can support patients receiving recombinant human growth hormone (r-hGH) treatment. Our focus is on examining adherence patterns during the first 6 weeks of treatment, with a particular emphasis on the days of the week, the variability in injection timings and the summer holiday. Our aim is to assess whether these early patterns can serve as indicators for predicting adherence in the subsequent 7-12 weeks of treatment. Data and Methods: Adherence data between March-November 2023 were extracted from the newly launched third generation of easypod™ device (EP3) and GrowZen Connect Next ecosystem. EP3 is the only connected device that delivers r-hGH and monitors real-time adherence to therapy. EP3 was perceived by health care professionals as more intuitive, comfortable, user-friendly, simpler, and easier to use than previous EP2. Patients with age 2-18 years at treatment start, a 7-day regimen and complete adherence data available between 1-12 weeks of treatment were selected. Data about sex, age at start treatment, as well as adherence, classified as high (≥95%) versus low/medium <95%), standard deviation (SD) of adherence (%), log (to adjust for skewness) of the interquartile range (IQR) of the timing of injection throughout the day, median timing of the injection after midnight, and ≥50% versus <50% of the injections fall in the summer period (July-August) aggregated by day over the first 6 weeks of treatment were included in a generalized linear mixed model with adherence between 7-12 weeks as outcome. Results: In total, data for 232 patients (124 boys and 108 girls, median (Q25-Q75) age at treatment start was 10.5 (7.6-12.3)) were available. The model showed no difference in effect of adherence rates during Monday-Thursday. We, therefore, aggregated the data by weekday (Monday-Thursday) versus weekend day (Friday-Sunday, Friday was included due to the common practice of administering injections during the evening). Adherence, SD of adherence, IQR of timing of injections, and injections falling in summer were significant (p<0.05) predictors of adherence during a weekday and weekend between 7-12 weeks. Areas under the curves were both 0.96 during the weekday and weekend. Conclusions: Our research shows that adherence and timing of injections, as well as taking into account the summer holidays, can predict future adherence accurately. Visual presentations showing aggregated adherence rates and variation of timing of injections by days of the week, as well as highlighting the summer periods may improve adherence risk management and timely support. Presentation: 6/3/2024

6818 Adherence Risk Management in Patients on Recombinant Human Growth Hormone Therapy, With Traffic Light Alerting Systems Tailored for Both Weekday and Weekend Adherence

Abstract Disclosure: S. Loche: Advisory Board Member; Self; SL has received advisory board fees from Merck KGaA, Darmstadt, Germany. Consulting Fee; Self; SL has received consultancy fees from Merck KGaA, Darmstadt, Germany. Speaker; Self; SL has received lecture fees from Merck KGaA, Darmstadt, Germany. P. van Dommelen: Consulting Fee; Self; PvD has a consultancy agreement with Merck KGaA, Darmstadt, Germany. V. Tornincasa: Employee; Self; VT is an employee of Ares Trading SA, an affiliate of Merck KGaA. L. Arnaud: Employee; Self; LA is an employee of Ares Trading SA, an affiliate of Merck KGaA. E. Koledova: Employee; Self; EK is an employee of Merck KGaA,Darmstadt, Germany. Stock Owner; Self; EK is an employee of Merck KGaA,Darmstadt, Germany. Background: The integration of data-driven strategies, such as clustering analysis with an alerting system, empowers healthcare professionals with actionable insights. In this study, we focused on clustering adherence patterns during the weekdays and weekends in the first 12 weeks of treatment. Our aim is to develop a data-driven clinical decision support system based on "traffic light" visualizations addressing adherence challenges during both weekdays and weekends in patients receiving Recombinant Human Growth Hormone Therapy. Data and Methods: Adherence data between March-November 2023 were extracted from the newly launched third generation of easypod™ device (EP3) and GrowZen Connect Next ecosystem. EP3 is the only connected device that delivers r-hGH and monitors real-time adherence to therapy. EP3 was perceived by health care professionals as more intuitive, comfortable, user-friendly, simpler, and easier to use than previous EP2. Patients with age 2-18 years at treatment start, a 7-day regimen and complete adherence data available between 1-12 weeks of treatment were selected. Patients’ adherence to therapy was represented using mean and standard deviation (SD) of adherence (%) and the interquartile range (IQR) of the timing of injections throughout the day in minutes aggregated by weekday (Monday-Thursday) and weekend day (Friday-Sunday). Cluster analysis was used to categorize adherence patterns using a Gaussian mixture model. Following a traffic lights-inspired visualization approach, the algorithm was set to create three clusters: high (green), medium (yellow), and low adherence (red). The area under the receiver operating characteristic curve (AUC-ROC) was used to find optimum thresholds for independent traffic lights. Results: Data for 232 patients (124 boys and 108 girls, median (Q25-Q75) age at treatment start was 10.5 (7.6-12.3)) were available. The most appropriate traffic light used the SD of adherence during the weekend, with an AUC-ROC value of 0.87. For the weekday, adherence-based traffic lights using optimum thresholds were >98% (SD<10.2), 88-98% (SD:10.2-22.8) and <88% (SD>22.8) for high, medium, and low adherence, respectively. For IQR of the timing of injections throughout the weekday, optimum thresholds were <38, 38-70 and >70 (min). For the weekend, optimum thresholds were >96% (SD<7.7), 86-96% (SD:7.7-21.5) and <86% (SD>21.5). For IQR of the timing of injections throughout the weekend, optimum thresholds were <56, 56-84 and >84 (min). Conclusions: Our research indicates that implementing a practical data-driven alert system, utilizing our proposed traffic-light coding, would empower healthcare practitioners to monitor patients' adherence to growth hormone therapy comprehensively. By understanding patient behavior during the weekdays and weekend, healthcare practitioners can tailor interventions for improved treatment outcomes. Presentation: 6/3/2024

Sustainable business operations in technology startups: A model for leadership and administrative excellence

Sustainable business operations have become a critical factor for success in technology startups, where rapid growth and resource constraints pose unique challenges. This review proposes a comprehensive model for sustainable operations management in tech startups, emphasizing the synergy between leadership and administrative excellence. The proposed model aims to guide startups in balancing profitability, scalability, and environmental responsibility while fostering long-term growth. The explores the challenges of achieving sustainability in startups, including limited resources, pressures of rapid scaling, and the environmental impact of tech-driven innovations. It emphasizes the pivotal role of visionary leadership in setting long-term sustainable goals, fostering a culture of innovation, and effectively managing change. Leadership’s ability to align sustainability with operational goals is critical in navigating the dynamic and competitive nature of the tech industry. Additionally, the review highlights the importance of administrative innovations in optimizing resource management, implementing data-driven decision-making, and streamlining processes. By leveraging automation and technology, startups can create more efficient and environmentally friendly operational systems. The model integrates leadership vision with administrative processes, creating a holistic approach that encourages continuous improvement and innovation. The benefits of adopting sustainable operations, such as enhanced financial resilience, competitive advantage, and improved brand reputation, are also discussed. Moreover, sustainability offers tech startups a significant edge in attracting and retaining top talent while fostering customer loyalty. Ultimately, this model serves as a blueprint for technology startups to build scalable, efficient, and responsible operations, ensuring their success in the long term by aligning leadership and administrative strategies with sustainability objectives. Keywords: Sustainable Business, Technology Startups, Leadership, Administrative Excellence.