Data-driven Process Research Articles

Design and Optimization of Heat Dissipation for a High-Voltage Control Box in Energy Storage Systems

Abstract To address the issue of excessive temperature rises within the field of electronic device cooling, this study adopts a multi-parameter optimization method. The primary objective is to explore and realize the design optimization of the shell structure of the high-voltage control box, aiming to effectively mitigate the temperature rise in internal components and enhance their thermal management efficacy without altering the fan performance, environmental conditions, or spatial layout. Initially, the study employs computational fluid dynamics methods to investigate the heat dissipation characteristics of the high-voltage control box, subsequently verifying the simulation parameters' accuracy through temperature rise tests. Building upon this foundation, the article conducts a thorough analysis of how the position and shape of the box's openings impact the device's temperature rise. The findings suggest that configuring circular openings on the front and rear sides can optimize the heat dissipation effect. Moreover, the SHERPA algorithm was employed to refine the size and distribution of the openings on the outer shell of the high-voltage control box through multi-parameter optimization, yielding locally optimal structural parameters. Post-optimization, the temperature measurement points within the high-voltage control box exhibited a maximum reduction in temperature rise of 27.16%. The pivotal contribution of this methodology is the application of a data-driven decision-making process for the enhancement of conventional heat dissipation designs. This research offers invaluable practical insights and novel perspectives on the optimization of thermal management designs for box-type electronic devices, significantly advancing the field of thermal management technology in electronic devices.

Empowering Large Language Models in Hybrid Intelligence Systems through Data-Centric Process Models

Hybrid intelligence systems aim to leverage synergies in closely collaborating teams of humans and artificial intelligence (AI). To guide the realization of such teams, recent research proposed design patterns that capture role-based knowledge on human-AI collaborations. Building on these patterns requires hybrid intelligence systems to provide mechanisms that orchestrate human and AI contributions accordingly. So far, it is unclear if such mechanisms can be provided based on shared representations of the required knowledge. In this regard, we expect ontology-based data-centric process modeling to be a promising direction for hybrid intelligence systems that aim to support knowledge-intensive processes (KiPs). We illustrate this through exemplary process models (realized with our ontology- and data-driven business process model -- ODD-BP) that reflect the team design patterns for hybrid intelligence systems. We point out that relying on such process models enables multiple actors to fulfill roles jointly and allows them to address individual shortcomings. This is examined by discussing integrating large language models (LLMs) into the process models and describing how complementary AI actors could help to empower LLMs to fulfill their role in human-AI collaboration more comprehensively. Future work will extend the provided concepts while their evaluation initially focuses on the KiP of medical emergency call handling.

A lightweight sensor ontology for supporting sensor selection, deployment, and data processing in forming processes

In the era of smart manufacturing, modern manufacturing systems face high demands for enhancing process performance and reducing machine downtime. Sensors and process data are essential for successfully implementing data-driven approaches to guarantee robust and reliable process monitoring, tool conditioning, or quality assurance. However, the accuracy and performance of such approaches are highly dependent on the quality of the gathered sensor data and influenced by the implemented data acquisition and processing methods. For this purpose, this work proposes a lightweight sensor ontology to provide a comprehensive overview to characterize underlying relationships between the physical environment and the quality of the data sets. The extended sensor ontology, in combination with domain knowledge, aims to support engineers in fully exploiting the potential of sensor data to obtain trustworthy data sets in forming technologies. As a result, this approach can improve the implementation of automated and data-driven process monitoring of forming systems and tools.

Semi-supervised process monitoring based on self-training PCA model

Data-centric engineering has recently become as a hot spot in both areas of artificial intelligence and data science, which shifts the focus of engineering application from model to data itself. The model for data-driven process monitoring is usually developed upon a large number of normal data samples, which have been assumed to be well evaluated in advance. However, the actual quality of those training data samples has a great impact to the monitoring performance. If those samples with low quality are included for modeling, the monitoring performance could be severely deteriorated. In practice, due to expensive and time-consuming data quality evaluation procedures, we may only have a quite limited number of evaluated data samples, and hold a large number of unevaluated data samples. This paper aims to develop a semi-supervised monitoring model, which can simultaneously incorporate the evaluated dataset and the original unevaluated dataset. As a result, the new semi-supervised monitoring scheme can save a lot of human efforts, and could be particularly useful in those processes which can only provide a small portion of evaluated data samples in time. The feasibility and efficiency of the new proposed monitoring scheme are examined through case studies of a numerical example and the TE benchmark process.

AI ENABLED WATER CONSERVATION FOR IRRIGATION USING IOT

This project introduces a smart irrigation system that utilizes Internet of Things (IoT) technology and machine learning algorithms to enhance water management in agriculture. The system employs a series of IoT sensors distributed across the field to consistently monitor key environmental factors such as soil moisture levels, temperature, humidity, and others. The collected data is analysed using machine learning Cat Boosting algorithm. This algorithm analyze the data to determine the optimal irrigation schedule based on crop water requirements, soil conditions, weather forecasts, and historical data. The system controls irrigation equipment such as pumps, valves, and sprinklers to deliver precise amounts of water to crops at the right time. Continuous feedback from the system allows for refinement of irrigation schedules, leading to improved water conservation, increased crop yield, and cost savings for farmers. This project discusses the benefits of such a system, including water conservation, increased crop yield, cost savings, environmental sustainability, and remote monitoring and control capabilities. Overall, the integration of IoT and machine learning technologies offers a powerful solution for sustainable agriculture, enabling data-driven decision-making processes to optimize water usage and maximize crop productivity. Keywords— Smart irrigation system, Internet of Things(IOT)technology, Water conservation, Environmental parameters, Soil moisture monitoring, Increased crop yield, Machine learning algorithms

No more war (for talent): the impact of HR analytics on talent management activities

PurposeDespite talent management’s (TM) importance for improving organizations' competitiveness and resilience, the pandemic highlighted the weakness of organizational-level TM strategies. Therefore, the objective of this study is to investigate the moderating impact of HR analytics on the relationship between TM and its individual outcomes (talent motivation and quality of hires) and subsequently, their impact on organizational outcomes (talent retention).Design/methodology/approachThe structural equation modeling (SEM) technique was used to analyze 219 online questionnaires administered to HR managers from European companies.FindingsA positive relationship exists between TM activities and talent motivation as well as the quality of hires. Furthermore, HR analytics positively moderates these relationships. Finally, talent motivation and the quality of hires are positively related to talent retention.Research limitations/implicationsThis study offers several contributions to theory, as it analyzes TM from an individual perspective and provides further empirical confirmation of the potential benefits of HR analytics and additional grounding to the contingency theory.Practical implicationsOur results will allow practitioners to better orient their HR investments, with positive effects for their organizations and their employees.Social implicationsThis study demonstrates that HR analytics can help organizations adopt a human-centric approach to TM, thus increasing the chances for talents to fully express their potential.Originality/valueThis study takes a step forward toward considering TM outcomes from an individual perspective, responding to new generations' need to pay more attention to their individualities. HR analytics can be a suitable tool to do so, as it can provide insights and suggestions based on the actual organizational context, making TM a more data-driven process.

Democratizing Data Visualization and Insights Extraction with Pandas, Generative AI, and CSV Data

Data visualization and insights extraction are crucial components of modern data-driven decision-making processes. However, traditional methods often require extensive coding knowledge, creating barriers for non-technical users. This whitepaper presents a comprehensive solution that integrates the powerful data manipulation capabilities of the Pandas library with cutting-edge Generative AI and natural language processing techniques. By leveraging a fine-tuned GPT-3 model trained on a diverse corpus of data analysis and visualization resources, our approach enables users to upload CSV data files and receive automated insights, default visualizations, and the ability to generate custom visualizations through intuitive natural language prompts. The solution streamlines the workflow, eliminating the need for coding expertise while ensuring data privacy and integrity within a secure execution environment. User studies and benchmarking demonstrate increased productivity, time savings, and high user satisfaction. This solution has the potential to democratize data analysis and visualization, empowering decision-makers across various industries with data-driven insights and informed decision-making processes.

Assessing indoor positioning system: A q-spherical fuzzy rough TOPSIS analysis

This study investigates advanced data collection methodologies and their implications for understanding employee and customer behavior within specific locations. Employing a comprehensive multi-criteria decision-making framework, we evaluate various technologies based on four distinct criteria and four technological alternatives. To identify the most effective technological solution, we employ the q-spherical fuzzy rough TOPSIS method, integrating three key parameters: lower set approximation, upper set approximation, and parameter q (where q ≥ 1). Our novel approach combines the TOPSIS method with q-spherical fuzzy rough set theory, providing deeper insights into data-driven decision-making processes in corporate environments. By comparing our proposed framework with existing multi-criteria decision-making methodologies, we demonstrate its strength and competitiveness. This research contributes to enhancing decision-making capabilities in corporate settings and beyond.

D³IKIT: data-driven design innovation kit

AbstractThe utilization of data in design is a crucial aspect of shaping the product and service development. Despite the lack of extensive research on this subject, this study aims to bridge the gap by introducing the ‘D³IKIT’, a data-driven design process and toolkit. Through workshops, this process and toolkit offer a practical method for creating innovative product and service concepts using data and machine learning. Developed and tested with the participation of 42 individuals, the ‘D³IKIT’ provides valuable insights for both practitioners and academics.

Anomaly Detection for Asynchronous Multivariate Time Series of Nuclear Power Plants Using a Temporal-Spatial Transformer.

Industrial process monitoring is a critical application of multivariate time-series (MTS) anomaly detection, especially crucial for safety-critical systems such as nuclear power plants (NPPs). However, some current data-driven process monitoring approaches may not fully capitalize on the temporal-spatial correlations inherent in operational MTS data. Particularly, asynchronous time-lagged correlations may exist among variables in actual NPPs, which further complicates this challenge. In this work, a reconstruction-based MTS anomaly detection approach based on a temporal-spatial transformer is proposed. It employs a two-stage temporal-spatial attention mechanism combined with a multi-scale strategy to learn the dependencies within normal operational data at various scales, thereby facilitating the extraction of temporal-spatial correlations from asynchronous MTS. Experiments on simulated datasets and real NPP datasets demonstrate that the proposed model possesses stronger feature learning capabilities, as evidenced by its improved performance in signal reconstruction and anomaly detection for asynchronous MTS data. Moreover, the proposed TS-Trans model enables earlier detection of anomalous events, which holds significant importance for enhancing operational safety and reducing potential losses in NPPs.

Privacy-Preserving data analysis

With the ever-increasing volume of data being generated and shared across various platforms, the challenge of maintaining privacy while extracting value from this data has become paramount. This paper delves into the realm of Privacy-Preserving Data Analysis (PPDA), examining its current landscape and the pivotal techniques shaping it. Using datasets from diverse domains, we evaluated four leading PPDA techniquesDifferential Privacy, Homomorphic Encryption, Secure Multi-Party Computation (SMPC), and Data Obfuscationto discern their efficacy and trade-offs in terms of data utility and privacy breach risk. Our findings underscore the strengths and constraints of each method, guiding researchers and practitioners in choosing the optimal approach for specific scenarios. As data continues to be an invaluable asset in the digital age, the tools and techniques to analyze it privately will play a critical role in shaping future data-driven decision-making processes.

EMPLOYEE PERFORMANCE ESTIMATION USING MACHINE LEARNING

In this research, we explore how machine learning is changing the way traditional HR practices handle employee reviews. The need for better evaluation methods for employees is increasing, and machine learning provides a variety of tools for analyzing data to meet this need. These algorithms can offer comprehensive and impartial assessments of performance, covering everything from productivity metrics to project outcomes and qualitative feedback. This study looks at the pros and cons of integrating this technology into HR practices, with the aim of enhancing the accuracy, fairness, and effectiveness of evaluations. The goal is to find a middle ground between technology and human judgment to improve performance reviews in modern workplaces. In today's fast-paced work environment, where excellence is key, traditional subjective performance assessments are becoming outdated. This research delves into the promising role of artificial intelligence in human resource management, particularly in creating detailed and fair performance evaluations. Picture AI algorithms as skilled musicians, harmonizing various data points – from productivity metrics to project achievements to qualitative input – to create a comprehensive picture of each employee's contributions. However, navigating this data-driven process can be complex. The focus of this study is on examining the positives and negatives of utilizing machine learning in HRM. The goal is to improve performance evaluations, ensuring they are both precise and unbiased. By connecting technology with a deeper human insight, organizations can move towards a future where performance appraisals truly maximize the abilities of their employees. Key Words: Performance, Evaluation, Employee Metrics, Human Resource Management (HRM), Examining

Development of a convolutional neural network to accurately detect land use and land cover

The detection and modeling of Land Use and Land Cover (LULC) play pivotal roles in natural resource management, environmental modeling and assessment, and ecological connectivity management. However, addressing LULCC detection and modeling constitutes a complex data-driven process. In the present study, a Convolutional Neural Network (CNN) is employed due to its great potential in image classification. The development of these tools applies the deep learning method. A methodology has been developed that classifies the set of land uses in a natural area of special protection. This study area covers the Sierra del Cando (Galicia, northwest Spain), considered by the European Union as a Site of Community Interest and integrated in the Natura 2000 Network. The results of the CNN model developed show an accuracy of 91 % on training dataset and 88 % on test dataset. In addition, the model was tested on images of the study area, both from Sentinel-2 and PNOA. Despite some confusion especially in the residential class due to the characteristics in this area, CNNs prove to be a powerful classification tool.•Classifications based on a CNN model•LULC are classified into 10 different classes•Training and test accuracy are 91 % and 88 %, respectively

Research and Practice of Online Postgraduate Quality Management Systems in Local Institutions with the Context of Digital Network

In the digital era, there is an urgent need to reconstruct the talent cultivation system with the support of computer technology, and it is crucial to create and build a quality control system for postgraduate education that meets the development of the times. The system is usually operated online using digital network technology. Most of the postgraduates admitted to the research group come from the same level of institutions or non-graduate training authorization units, and their scientific research ability is relatively weak, hence they need more guidance and help from the university and supervisors in the cultivation process. Relying on the university's "Intelligence +" academic characteristics, the group has explored and practiced digital transformation in terms of accurate publicity, quality improvement, and differentiation cultivation, which mainly includes the following two aspects. The first, take the online teaching management platform as the core, cooperate with the enrollment platform, the practice results platform, and the graduation work platform through the network, with the user experience of teachers and students as the test standard, mine data on the habits of teachers and students, realize the orderly operation of the data-driven postgraduate education process. The second is to improve the teaching management ability and innovation consciousness of teachers under the intelligence education conditions through data platform and online or underline staged training. At the same time, the Group will continue to pay attention to data security and privacy protection issues in the postgraduate QMS, and deeply explore the integration and application of big data technology and education management, so as to improve the management level of postgraduate education and adapt to the new situation and requirements of postgraduate education.

HARNESSING DATA INSIGHTS FOR CRISIS MANAGEMENT IN U.S. PUBLIC HEALTH: LESSONS LEARNED AND FUTURE DIRECTIONS

The effective management of crises in U.S. public health relies heavily on harnessing insights from data. This paper examines the lessons learned from past crises and outlines future directions for leveraging data to enhance crisis management efforts. Drawing on case studies of significant events such as the COVID-19 pandemic and natural disasters, we analyze the role of data in informing decision-making, identifying challenges, and deriving best practices. Despite the advancements in data infrastructure, including real-time monitoring systems and predictive analytics, there remain significant challenges in data collection, integration, and analysis. The paper highlights the importance of improving data governance, investing in technology, and fostering interdisciplinary collaboration to address these challenges. Additionally, considerations such as equity and privacy are crucial in the development and implementation of data-driven strategies. Looking ahead, the paper provides recommendations for enhancing crisis management practices, including the adoption of standardized data protocols, the development of early warning systems, and the promotion of data-driven decision-making processes. By prioritizing data-driven approaches and embracing continuous learning, stakeholders can better prepare for and respond to future public health crises. This paper serves as a roadmap for policymakers, public health officials, and researchers to optimize data utilization in crisis management, ultimately safeguarding the well-being of communities across the United States.
 Keywords: Data Insights, Crisis Management, in U.S. Public Health, Lessons Learned, Future Directions.

Empowering Smart Cities with AI and RPA: Strategies for Intelligent Urban Management and Sustainable Development

This research explores the transformative potential of Artificial Intelligence (AI) and Robotic Process Automation (RPA) in empowering smart cities to achieve intelligent urban management and sustainable development. Through a comprehensive analysis of literature, case studies, and qualitative research methods, the paper identifies key strategies for leveraging AI and RPA to address urban challenges and promote sustainable urban development. The integration of AI and RPA technologies enables data-driven decision-making processes, streamlines administrative workflows, and enhances service delivery in smart cities. Furthermore, AI and RPA contribute to promoting sustainable development goals by optimizing resource utilization, improving environmental management practices, and enhancing resilience to climate change. However, the widespread adoption of AI and RPA in smart cities faces challenges related to privacy, data security, and equity, which must be carefully addressed to ensure responsible and equitable deployment of these technologies. By adopting comprehensive strategies, fostering collaboration between stakeholders, and embracing a culture of innovation, cities can harness the full potential of AI and RPA to build smarter, more resilient, and sustainable urban environments for all residents. This research provides valuable insights for policymakers, urban planners, and technology providers seeking to leverage AI and RPA to address urban challenges and promote sustainable development in smart cities.

Hybrid Intelligence Systems Combining Human Expertise and AI/RPA for Complex Problem Solving

Hybrid Intelligence Systems (HIS) represent a paradigm shift in problem-solving methodologies by integrating human expertise with Artificial Intelligence (AI) and Robotic Process Automation (RPA). This paper explores the mechanisms, applications, benefits, challenges, and future directions of HIS in the context of complex problem-solving. Through collaborative synergies between human cognition and machine intelligence, HIS enhances decision-making accuracy, efficiency, and innovation. Human experts contribute domain knowledge, contextual understanding, and ethical reasoning, while AI algorithms and RPA systems offer data-driven insights, computational power, and process automation capabilities. HIS fosters inclusivity, diversity, and democratization in problem-solving processes by harnessing the collective intelligence of diverse teams and stimulating interdisciplinary collaboration. However, challenges such as privacy concerns, data security risks, and algorithmic biases must be addressed to realize the full potential of HIS. Looking ahead, the integration of Explainable AI (XAI), Edge AI, and Neuro symbolic AI holds promise for enhancing transparency, interpretability, and robustness in HIS architectures. Human-centered design principles and interdisciplinary research collaborations will shape the development and deployment of HIS, ensuring alignment with human values, preferences, and needs. Ultimately, HIS will continue to serve as a beacon of collaboration, creativity, and collective intelligence in shaping a better world for generations to come.

Data-driven determination of zooplankton bioregions and robustness analysis

Identifying biogeographic regions through cluster analysis of species distribution data is a common method for partitioning ecosystems. Selecting the appropriate cluster analysis method requires a comparison of multiple algorithms. In this study, we demonstrate a data-driven process to select a method for bioregionalization based on community data and test its robustness to data variability following these steps:•We aggregated and curated zooplankton community observations from expeditions in the Northeast Pacific.•We determined the best bioregionalization approach by comparing nine cluster analysis methods using ten goodness of clustering indices.•We evaluated the robustness of the bioregionalization to different sources of sampling and taxonomic variability by comparing the bioregionalization of the overall dataset with bioregionalizations of subsets of the data.The K-means clustering of the log-chord transformed abundance was selected as the optimal method for bioregionalization of the zooplankton dataset. This clustering resulted in the emergence of four bioregions along the cross-shelf gradient: the Offshore, Deep Shelf, Nearshore, and Deep Fjord bioregions. The robustness analyses demonstrated that the bioregionalization was consistent despite variability in the spatial and temporal frequency of sampling, sampling methodology, and taxonomic coverage.

Integrated application of synergetic approach for enhancing intelligent steam generator control systems

This article focuses on the integrated application of the synergetic approach to enhance the quality of intelligent steam generator control systems.By combining various techniques such as model-based control, adaptive control, and artificial intelligence, an efficient and flexible control system can be developed. Model-based control utilizes mathematical models of steam generators to formulate control algorithms and predict system behavior. Adaptive control enables the system to adapt to changing conditions by adjusting control parameters based on real-time measurements. Artificial intelligence techniques, including neural networks and genetic algorithms, facilitate learning, optimization, and data-driven decision-making processes. The objectives of this research are to investigate the benefits of the synergetic approach in steam generator control, including improved steam generation efficiency, optimized energy consumption, enhanced system stability and reliability, and adaptability to varying operating conditions and disturbances. The findings and conclusions of this study are expected to provide valuable insights for engineers, researchers, and professionals involved in the design and implementation of intelligent steam generator control systems. By integrating the synergetic approach, substantial enhancements in control quality can be achieved, leading to optimal operation and maximum efficiency of power plants.

Integration of Large Language Models with IoT in Smart Agriculture to Improve Efficiency, Yield, and Quality

This paper examines the novel integration of large language models (LLMs) with Internet of Things (IoT) technologies in the context of smart agriculture. The study emphasizes the transformative impact of LLMs in enhancing data-driven decision-making processes through advanced natural language processing capabilities. By employing IoT devices, comprehensive real-time data on environmental conditions, soil health, and crop status are collected and analyzed using LLMs, thereby facilitating the generation of actionable insights for precision agriculture. Key areas of improvement include optimized irrigation scheduling, targeted pest and disease management, and efficient resource utilization, which collectively contribute to increased crop yields and quality. The findings illustrate the potential of combining LLMs with IoT to create a sustainable, efficient, and high-yield agricultural ecosystem.