Data Quality Requirements Research Articles

Machine Learning in Cybersecurity: A Comprehensive Analysis of Intrusion Detection Systems

Cyber threats are getting more complex and larger, and traditional security measures can't keep up with the sophisticated attacks. Machine Learning (ML) has emerged as a powerful tool in enhancing cybersecurity, particularly in the development of Intrusion Detection Systems (IDS). a comprehensive analysis of the role of machine learning in intrusion detection, focusing on its ability to detect and respond to both known and unknown threats in real-time. various machine learning techniques, including supervised, unsupervised, and deep learning models, and evaluates their effectiveness in identifying anomalies and preventing security breaches. Key performance metrics such as accuracy, false positive rates, and scalability are analyzed, with a focus on the advantages and limitations of each approach. Additionally, the paper explores the challenges associated with implementing ML-based IDS, including data quality, adversarial attacks, and computational requirements. Through case studies and real-world examples, this paper highlights how machine learning is revolutionizing intrusion detection and presents insights into future advancements that will further strengthen cybersecurity defenses.

TECHNOLOGY AND ARTIFICIAL INTELLIGENCE'S IMPACT ON ESG REPORTING QUALITY

When it comes to communicating a company's performance in terms of sustainability, ESG reporting, which is an acronym that stands for environmental, social, and governance reporting, is a key component. On the other hand, there are still issues with standards, data quality, and consistency requirements. The purpose of this study is to investigate the ways in which digital technology and artificial intelligence (AI) have the potential to enhance environmental, social, and governance (ESG) reporting by resolving these issues. For the purpose of streamlining data collecting, improving quality, and simplifying communication, digital technology and artificial intelligence employ a broad variety of complex approaches. Some examples of these methods include blockchain technology, data analytics, machine learning, and natural language processing. Using regression analysis and data collected from A-share listed companies between the years 2012 and 2021, this study investigates the relationship between digital transformation and the environmental, social, and governance (ESG) performance of businesses. The study has an emphasis on the repercussions that are associated with artificial intelligence, digital technology, and ESG reporting (environmental, social, and governance). A number of different aspects, including corporate governance, risk assessment, shareholder value, and corporate responsibility, are all affected by these ramifications. For any business that aspires to grow, develop, become more efficient, and maintain a competitive advantage over its rivals, digital technology and artificial intelligence are indispensable tools. The reporting of environmental, social, and governance (ESG) issues contributes to the advancement of sustainable development and has a beneficial impact on the risk profile, reputation, performance, and total value of entities. As a result of a comprehensive study, methodological inquiry, and analysis, this research provides useful insights and ideas for enhancing environmental, social, and governance (ESG) reporting via the use of digital technology and artificial intelligence. It is discovered that the application of digital technology and artificial intelligence results in ESG reporting that is more comprehensive and has a better degree of confidence.

The performance of alternative GPS tracking devices: a case report on wild boars (Sus scrofa)

BackgroundTelemetry studies on terrestrial mammals have been carried out almost exclusively using neck collar-mounted devices. However, collars are not suitable for all species and all age classes and may therefore compromise animal welfare and data quality requirements. Here, we evaluate the use of non-collar tracking devices on wild boars (Sus scrofa), a species for which collars may be problematic due to neck shape and seasonal body mass variation. We tested pelt-glued tags and three types of GPS ear tags. In addition, animals were marked with plain cattle ear tags.ResultsThe tested GPS tags exhibited various technical challenges related mostly to satellite coverage of the study area and tag loss. However, GPS devices enabled moderate movement monitoring of weeks to months, whereas plain cattle ear tags offered information on displacement over longer time intervals. The longest direct dispersal distance (163 km) by a wild boar sounder was discovered by cattle tags.ConclusionsData volume and quality obtained from small tags are inferior to data provided by GPS collars, but low-fix rate tags may enable monitoring of individuals with reduced invasiveness. Moreover, these tags enable tracking in cases where the collar is not an option. The low price and small size of the devices together with technical developments may offer cost-effective tools for future studies on dispersal and survival.

Quantifying the impact of flexibility asset location on services in the distribution grid: Power system and local flexibility market co-simulation

This research investigates the effectiveness of incorporating locational sensitivity factors into local flexibility market clearing mechanisms for effective congestion management and voltage regulation in distribution grids. A centralized local flexibility market optimization model is developed that considers technical and economic constraints. The study aims to explore the requirements for data availability, data quality, and reliable data exchange that can facilitate a broader range of flexibility services, thereby promoting the development of a local flexibility market. Sensitivity factors, including power transfer distribution factors, voltage sensitivity coefficients and transformer sensitivity coefficients, are used to quantify the impact of flexibility asset locations on congestion management and inform clearing rules. These static metrics are insufficient for establishing effective local flexibility market clearing rules when flexibility is procured by distribution system operators. The approach considers the state of the grid when calculating the sensitivity coefficients, which leads to a more accurate evaluation of flexibility bids, especially with regard to the impact of location on congestion management. The proposed mechanism for clearing the local flexibility market assumes continuous communication between the proposed local flexibility market operator and the distribution system operator for dynamic, iterative market clearing, which ensures the protection of grid data and a more accurate bid evaluation. The study demonstrates that the inclusion of locational information significantly increases the effectiveness of the proposed local flexibility market-based congestion management. The developed simulator for the proposed local flexibility market provides valuable insights into the interaction between the proposed local flexibility market and the distribution grid. The research results, derived from selected use cases, emphasize the importance of location-based sensitivity factors in the proposed local flexibility market clearing for distribution grids. The proposed approach offers a promising solution for optimizing congestion management and voltage regulation while ensuring efficient integration of distributed energy resources into distribution grids.

Data quality-oriented scan planning for steel structure scenes using a probabilistic genetic algorithm

Scan planning is often challenging particularly in steel structure scenes because of its complex shapes and occlusions. Meeting the requirements of data quality for the scan-to-BIM model is also another issue for accurate point cloud data acquisition. To address these issues, this study proposes a solution that determines an optimal number of scans and corresponding scan positions and parameters. Three primary steps include 1) extraction of feature points using a slicing cutting method and range images, 2) evaluation of data quality using visibility check and data density evaluation, and 3) determination of optimal scan configuration using a probabilistic genetic algorithm. In order to validate the proposed solution, a series of lab-scale experiments involving five case studies with different scenarios are conducted and the results show a similarity of 88.4% between simulation and actual experiments, demonstrating the feasibility of the proposed method for steel structure scenes with complex shapes and occlusions.

Application of Machine Learning in the Field of Diabetes

Diabetes, a prevalent chronic disease worldwide, poses significant challenges to healthcare systems, with its cases steadily increasing. Machine learning has emerged as a promising tool in diabetes research, offering opportunities for prediction and diagnosis through analysis of vast clinical datasets. This paper systematically reviews and analyzes the latest research progress in ML applications for diabetes, encompassing risk prediction and clinical diagnosis. Various ML techniques, including neural networks, supervised learning, semi-supervised learning, deep learning, and ensemble learning, are explored in their application to diabetes research. While ML has shown promise in improving diagnostic accuracy and efficiency, several challenges remain, such as data quality, model interpretability, and computational requirements. Recommendations for future research focus on addressing these challenges to further advance ML’s effectiveness in diabetes management. Through this review, we aim to provide valuable insights for researchers and clinicians, promoting the continued development and application of ML technology in diabetes care, improving efficiency and accuracy.

Technical noise, data quality, and calibration requirements for next-generation gravitational-wave science

Abstract The next generation of ground-based gravitational-wave interferometers is expected to generate a bounty of new astrophysical discoveries, with sensitivities and bandwidths greatly improved compared to current-generation detectors. 
These detectors will allow us to make exceptional advancements in our understanding of fundamental physics, the dynamics of dense matter, and the cosmic history of compact objects. 
The fundamental design aspects of these planned interferometers will enable these new discoveries; 
however, challenges in technical noise, data quality, and calibration have the potential to limit the scientific reach of these instruments.
In this work, we evaluate the requirements of these elements for next-generation gravitational-wave science, focusing on how these areas may impact the proposed Cosmic Explorer observatory.
We highlight multiple aspects of these fields where additional research and development is required to ensure Cosmic Explorer reaches its full potential.

Cognitive digital twin-enabled multi-robot collaborative manufacturing: Framework and approaches

Multi-robot collaborative manufacturing (MRCMfg) is vital to ensure effective and efficient operation in the manufacturing industry. In recent years, enabled by the new generation of information technology, represented by the digital twin (DT), the digitalization of MRCMfg has progressed steadily. However, more complex manufacturing tasks and more customized manufacturing demands pose challenges to MRCMfg, which limit the applicability of MRCMfg in practice. Recent successes in intelligent manufacturing show that empowering DT with the intelligence and cognition capabilities to adapt to the above challenges is a promising approach. Inspired by that, a cognitive digital twin (CDT) framework with multiple artificial intelligence algorithms for MRCMfg is proposed. The framework consists of three critical spaces: (i) Physical-virtual space: multi-agent system-based DT; (ii) Cognition space: multi-behavior tree (BT) cognition models; (iii) Data space: data-driven cognition evolution. In the data space, multimodal perception is a important way to improve the cognition capabilities of MRCMfg. To alleviate the multimodal data quality requirements as well as to improve the robustness of the perception model, an incomplete multimodal Transformer with deep autoencoder (IMTDAE) approach is proposed. The effectiveness and feasibility of this framework and IMTDAE are verified by case studies. The results provide compelling evidence that this framework has the potential to significantly enhance the intelligence and cognition capabilities of MRCMfg.

Applications of Finite-Time Lyapunov Exponent in detecting Lagrangian Coherent Structures for coastal ocean processes: a review

Addressing the threats of climate change, pollution, and overfishing to marine ecosystems necessitates a deeper understanding of coastal and oceanic fluid dynamics. Within this context, Lagrangian Coherent Structures (LCS) emerge as essential tools for elucidating the complexities of marine fluid dynamics. Methods used to detect LCS include geometric, probabilistic, cluster-based and braid-based approaches. Advancements have been made to employ Finite-time Lyapunov Exponents (FTLE) to detect LCS due to its high efficacy, reliability and simplicity. It has been proven that the FTLE approach has provided invaluable insights into complex oceanic phenomena like shear, confluence, eddy formations, and oceanic fronts, which also enhanced the understanding of tidal-/wind-driven processes. Additionally, FTLE-based LCS were crucial in identifying barriers to contaminant dispersion and assessing pollutant distribution, aiding environmental protection and marine pollution management. FTLE-based LCS has also contributed significantly to understanding ecological interactions and biodiversity in response to environmental issues. This review identifies pressing challenges and future directions of FTLE-based LCS. Among these are the influences of external factors such as river discharges, ice formations, and human activities on ocean currents, which complicate the analysis of ocean fluid dynamics. While 2D FTLE methods have proven effective, their limitations in capturing the full scope of oceanic phenomena, especially in 3D environments, are evident. The advent of 3D LCS analysis has marked progress, yet computational demands and data quality requirements pose significant hurdles. Moreover, LCS extracted from FTLE fields involves establishing an empirical threshold that introduces considerable variability due to human judgement. Future efforts should focus on enhancing computational techniques for 3D analyses, integrating FTLE and LCS into broader environmental models, and leveraging machine learning to standardize LCS detection.

A data quality management framework for equipment failure risk estimation: Application to the oil and gas industry

In engineering asset management, accurate failure risk estimation is essential for averting equipment breakdowns and optimizing risk-based maintenance strategies. Data quality and model fitting are two critical sources of prediction uncertainty in risk estimation. While much attention has been devoted to model fitting for risk estimation, the critical role of data quality has often been overshadowed. To bridge this research gap, this paper presents a novel data quality management framework tailored for industrial equipment failure risk estimation. The framework covers the steps from data to model. It consists of the following main phases: data development, data quality assessment, data quality requirement decision-making, data quality improvement, and risk estimation model development. The framework provides detailed guidelines that can facilitate data practitioners to build individualized data quality requirement decision-making models for failure risk estimation of their equipment. The decision-making model can measure the adequacy of existing data to build a risk estimation model that meets the specified requirements and further determines the best risk estimation model given the available data. A case study using actual data collected during oil well drilling operations from multiple oil fields demonstrates the practicality and effectiveness of the framework. In this case study, four risk estimation models are compared, including two baseline models (mean time to failure and median time to failure) and two machine-learning models (quantile regression and hidden Markov model). In addition, a decision tree-based decision model is developed to determine whether the data quality meets the requirements and the best risk estimation model in case the data quality meets the requirements.

GAN-BodyPose: Real-time 3D human body pose data key point detection and quality assessment assisted by generative adversarial network

With the rapid advancement of deep learning and computer vision, these technologies are becoming increasingly vital in areas like virtual reality, medical diagnosis, and sports training. Existing methods for real-time 3D human body pose keypoint detection and quality assessment face significant challenges such as insufficient detection accuracy, low computational efficiency, and high data quality requirements. To address these challenges, we propose an innovative solution, GAN-BodyPose. This approach integrates 3D convolutional neural networks, self-attention mechanisms, and generative adversarial networks to deliver efficient and accurate detection and assessment in real time. The GAN-BodyPose framework combines 3D-CNN and self-attention for effective feature extraction and keypoint detection, enhanced further by generative adversarial networks for superior data quality and accuracy. Our extensive evaluations using a large-scale 3D human body pose dataset demonstrated that GAN-BodyPose outperforms traditional methods, showing improvements in processing speed (15% faster), accuracy in terms of Mean Per Joint Position Error (reduced by approximately 2.2%), and an Area Under the Curve (AUC) score increased by approximately 9.5% compared to HR-Net and other datasets. Additionally, it achieves lower Floating-Point Operations (FLOPs) by about 9.3%, indicating more efficient computational performance. These advancements underline the potential of our approach to significantly enhance user experiences in virtual reality, motion capture, and other real-time applications. The successful application of GAN-BodyPose promises greater efficiency and precision in fields ranging from game development to medical diagnostics, and robust support for human-computer interaction and gesture recognition. This research represents a substantial contribution to deep learning applications in robot control, decision-making, and broadens the research foundation in these domains.

African time travellers: What can we learn from 500 years of written accounts?

AbstractIn this paper we study 500 years of African economic history using traveller accounts. We systematically collected 2464 unique documents, of which 855 pass language and rigorous data quality requirements. Our final corpus of texts contains more than 230 000 pages. Analysing such a corpus is an insurmountable task for traditional historians and would probably take a lifetime's work. Applying modern day computational linguistic techniques such as a structural topic model approach (STM) in combination with domain knowledge of African economic history, we analyse how first‐hand accounts (topics) evolve across space and time. Apart from obvious accounts of climate, geography, and zoology, we find topics around imperialism, diplomacy, conflict, trade/commerce, health/medicine, evangelization, and many more topics of interest to scholarship. We illustrate how this novel database and text analysis can be employed in three applications (1) What views are introduced by travellers as a result of their occupational background? (2) Did the adoption of quinine as treatment and prophylaxis against malaria facilitate European expansion into Africa? (3) When and how did the diffusion of New World crops alter the African economic landscape?

Optimizing Energy Efficiency in Smart Grids Using Machine Learning Algorithms: A Case Study in Electrical Engineering

The increasing demand for power driven by the integration of renewable energy sources has created an urgent need to improve energy efficiency in smart grids Conventional power grids with unidirectional power supply and midstream industries struggling to accommodate variable renewable energy and increased demand consumption. In response, this research explores the use of machine learning (ML) algorithms as a solution to increase energy efficiency, and presents a data-driven approach to address the challenges of modern smart grids meeting the solution for. The effectiveness of the algorithms is examined. Specifically, the research aims to (1) investigate how ML models can improve power delivery and reduce power consumption, (2) differ in key metrics such as accuracy and responsiveness, among others and regression, clustering, and neural networks -Time for performance testing ML algorithms, and (3) ML applications in smart networks Address practical challenges, such as data quality and computational requirements. To achieve these objectives, the study seeks to provide actionable insights for practitioners and researchers aiming to adopt ML solutions for sustainable energy use. The results of this study show that the ML algorithm significantly increases the energy consumption of smart grids. Through predictive modeling and optimization, the ML model achieved a 15% improvement in energy efficiency, a 25% reduction in peak demand, and an annual cost savings of approximately $200,000 Furthermore, predicting a ML-driven maintenance enabled early detection of potential grid issues, reducing downtime , technical losses and it has been reduced. These findings highlight the potential of ML to address today’s complex energy systems, delivering robust, scalable, and efficient solutions that support the integration and powering of renewable energy sources sustainable planning goals are advanced.

Generative Artificial Intelligence Enhancements for Reducing Image-based Training Data Requirements

ObjectiveTraining data fuel and shape the development of AI models. Intensive data requirements are a major bottleneck limiting the success of AI tools in sectors with inherently scarce data. In healthcare, training data is difficult to curate, triggering growing concerns that the current lack of access to healthcare by under-privileged social groups will translate into future bias in healthcare AIs. In this report, we developed an autoencoder to grow and enhance inherently scarce datasets to alleviate our dependence on big data. DesignComputational study with open-source data SubjectsThe data was obtained from six open-source datasets comprising patients aged 40 to 80 in Singapore, China, India, and Spain. MethodsThe reported framework generates synthetic images based on real-world patient imaging data. As a test case, we used autoencoder to expand publicly available training sets of optic disc photos, and evaluated the ability of the resultant datasets to train AI models in the detection of glaucomatous optic neuropathy.Main Outcome MeasuresAUC (Area Under the ROC Curve) were used to evaluate the performance of the glaucoma detector. A higher AUC indicates better detection performance. ResultsResults show that enhancing datasets with synthetic images generated by autoencoder led to superior training sets that improved the performance of AI models. ConclusionsOur findings here help address the increasingly untenable data volume and quality requirements for AI model development and have implications beyond healthcare, towards empowering AI adoption for all similarly data-challenged fields.

Data augmentation-based enhanced fingerprint recognition using deep convolutional generative adversarial network and diffusion models

The progress of fingerprint recognition applications encounters substantial hurdles due to privacy and security concerns, leading to limited fingerprint data availability and stringent data quality requirements. This article endeavors to tackle the challenges of data scarcity and data quality in fingerprint recognition by implementing data augmentation techniques. Specifically, this research employed two state-of-the-art generative models in the domain of deep learning, namely Deep Convolutional Generative Adversarial Network (DCGAN) and the Diffusion model, for fingerprint data augmentation. Generative Adversarial Network (GAN), as a popular generative model, effectively captures the features of sample images and learns the diversity of the sample images, thereby generating realistic and diverse images. DCGAN, as a variant model of traditional GAN, inherits the advantages of GAN while alleviating issues such as blurry images and mode collapse, resulting in improved performance. On the other hand, Diffusion, as one of the most popular generative models in recent years, exhibits outstanding image generation capabilities and surpasses traditional GAN in some image generation tasks. The experimental results demonstrate that both DCGAN and Diffusion can generate clear, high-quality fingerprint images, fulfilling the requirements of fingerprint data augmentation. Furthermore, through the comparison between DCGAN and Diffusion, it is concluded that the quality of fingerprint images generated by DCGAN is superior to the results of Diffusion, and DCGAN exhibits higher efficiency in both training and generating images compared to Diffusion.

Comparison-Transfer Learning Based State-of-Health Estimation for Lithium-Ion Battery

Abstract Rapid and accurate estimation of the state of health of lithium-ion batteries is of great significance. This paper aims to address two issues faced when applying deep learning methods to estimate the health status of lithium-ion batteries: high data quality requirements and poor model generalizability. This paper proposes a comparison-transfer learning approach with cyclic synchronization to estimate the state of health of lithium-ion batteries. First, a cyclic synchronization method based on the Bezier curve fitting algorithm is introduced to synchronize the data obtained at different charge–discharge cycles of the lithium-ion battery, facilitating input to the model. Second, a comparison-transfer network using the Pearson correlation coefficient is proposed to transfer knowledge from the source dataset to predict the target dataset under different environmental temperatures. By training a pre-trained model on the source dataset and utilizing the correlation coefficient to analyze the similarity between the source and target datasets, the accumulated knowledge in the source dataset can be effectively utilized to enhance prediction performance on the target dataset. In the experiments, the proposed method is validated using the lithium-ion battery aging public datasets. The experimental results demonstrate that the proposed approach achieves superior prediction performance in the case of small-sample sizes, exhibiting higher accuracy and stability compared to traditional deep learning methods.

Multi-criteria model for selection of optical linear terminals based on FUZZY TOPSIS method

Optical networks are an integral part of modern telecommunication systems. Huge traffic and ever-increasing requirements for data transmission capacity and quality encourage the wider use of modern optical technologies in telecommunications. The rational choice of equipment for the design of optical networks is an urgent task at present. The subject of this study is the process of selecting optical line terminals, which is associated with the evaluation of possible options by a set of indicators. The OLT (optical line terminal) in PON technology implements the function of organizing subscriber lines and is also a node equipment - an L4 switch that combines the functions of routing (IP), traffic fragmentation (VLAN), switching (MAC), quality of service (QoS), and some necessary network service functions. The optimal structuring of PON access networks demands a judicious selection of software and hardware, guided by their tactical and technical characteristics, in view of the substantial information load they entail. One of the effective approaches to solving such problems is the use of MCDM (Multiple Criteria Decision Making) methods. Purpose: to construct a TOPSIS model of the optimal choice of optical linear terminals in the conditions of unclear information for different cases of aggregation of evaluations of decision makers. Task: to formalize the process of selecting optical linear terminals; develop a multi-criteria mathematical model for the effective selection of optical line terminals. Methods used in the study: The fuzzy TOPSIS method, four methods of aggregating the opinions of decision makers. The following results were obtained. Alternatives and criteria for their evaluation are defined. On the basis of interviews with decision makers, evaluations of the degree of importance of the criteria and alternatives to the criteria were determined. Linguistic changes were used to describe decision-makers’ evaluations, which were interpreted as triangular fuzzy numbers. During this study, four methods of aggregating the evaluations of decision makers were used. A fuzzy TOPSIS model for selecting an optical line terminal is constructed. The ranking of the selected optical line terminals is obtained and the best alternative is determined. The results of modeling with different methods of aggregating assessments of decision makers are compared. Conclusions. The use of the Fuzzy TOPSIS method for optimal selection of optical linear terminals is proposed. The influence of the methods of aggregation of evaluations of decision-makers is analyzed.

A novel classification method for GPR B-scan images based on weak-shot learning

Ground Penetrating Radar (GPR) is extensively employed for underground target detection, and the advancements in this domain have been hastened by the utilization of deep learning methodologies. Despite advancements in deep learning, practical engineering applications often face challenges in meeting the demanding requirements of large-scale data and high data quality. In this study, based on cross-domain transfer learning, we proposed a method for underground target classification of weak labeled GPR data. It combined strongly annotated source domain data with weakly annotated target domain data through adversarial learning to leverage the prior knowledge from existing large datasets and addressed the classification problem of small-sample GPR data. By employing adversarial learning to integrate strongly labeled source domain data with weakly labeled target domain data, we leverage prior knowledge from existing large datasets, thereby enhancing classification accuracy on a small GPR dataset. This approach reduces the requirements for annotation quality and sample quantity in the GPR B-scan dataset of underground targets. Additionally, it exhibits robustness to label noise, making it applicable to real-world GPR data classification tasks. Compared to some well-known networks, the proposed model achieved the best classification performance in multiclass classification problems, with an accuracy of 87.62% and an F1macro score of 87.60% on the test set. These results represent an improvement of 17.14 and 17.20 percentage points, respectively, over the best-performing classical classification networks. Additionally, the proposed model's reliability in cross-domain transfer learning, even with limited data, was substantiated through a comparison with other few-shot learning networks.

QI$$^2$$: an interactive tool for data quality assurance

AbstractThe importance of high data quality is increasing with the growing impact and distribution of ML systems and big data. Also, the planned AI Act from the European commission defines challenging legal requirements for data quality especially for the market introduction of safety relevant ML systems. In this paper, we introduce a novel approach that supports the data quality assurance process of multiple data quality aspects. This approach enables the verification of quantitative data quality requirements. The concept and benefits are introduced and explained on small example data sets. How the method is applied is demonstrated on the well-known MNIST data set based an handwritten digits.

Data quality in environmental assessment methods – Implications for the operational management in manufacturing

To implement environmentally sustainable, resource-efficient manufacturing, the quantification of the environmental impacts is a central task. This task requires suitable methods. Various lifecycle phases, including remanufacturing concepts, and the associated increasing complexity demand a change from static to dynamic assessment methods. Flexible changes must be considered. Thereby, data quality has a major impact on the subsequent assessment results. Though the available data amount in a manufacturing environment is constantly increasing, the influence of data quality on the respective method – and thus, the reliability of the results – remains a key issue. To address this problem, this paper investigates the necessary developments regarding data quality for environmental assessment methods in dynamically changing manufacturing settings. First, relevant data quality dimensions are identified. Subsequently, a cross-comparison of data quality dimensions and different approaches of environmental evaluation in manufacturing is developed. The necessary changes to data quality requirements in operational environmental assessments are derived. Finally, a concept for the improvement of data quality in dynamic assessments is proposed and discussed based on the use case of battery modules’ remanufacturing. This paper contributes to the development of environmental assessment methods that are evolving from static, historical data-based evaluations to dynamic assessment tools of flexible productions.