Expert Rules Research Articles

Synthesis and mutagenic risk of avanafil's potential genotoxic impurities.

In the technical route for the synthesis of avanafil, 1-ethyl-(3-dimethylaminopropyl)carbamyldiimide hydrochloride (EDCI) and 1-hydroxybenzotriazole (HOBT) are used as reactive acid-amine binding agents. HOBT contains trace amounts of hydrazine residue, and there is a risk of introducing potentially mutagenic impurities with hydrazide-containing structures. The potentially genotoxic impurities E (Imp-E) and F (Imp-F) of avanafil with altering hydrazide-structure were synthesized by chemical method; subsequently, the impurities were evaluated and classified according to ICH M7 guidelines. Two complementary quantitative structure-activity relationship (QSAR) evaluation systems (Derek and Sarah) based on expert rules and statistics were used to preliminarily predict the genotoxicity of Imp-E and Imp-F, and the prediction result of E was suspected to be positive. In the Ames test of Imp-E and Imp-F, in the dose range of 62.5-1000 μg per plate, with or without the presence of metabolic activation system S9, the number of revertant colonies did not exceed 2 times the number of colonies in the solvent control group and did not show a dose-response relationship, and the test results were negative. Imp-E and Imp-F were determined to be negative for genotoxicity, which could be controlled as class 5 in ICH M7, that is, non mutagenic impurity.

THE BENEFITS OF USING DIGITAL TECHNOLOGY IN FORENSIC ACCOUNTING EXPERTISE

The article is devoted to the current problem of using modern digital technologies in the production of forensic financial and economic expertise. In the recent past, in the framework of criminal cases, experts had to study a significant array of accounting documents and other documents containing economically significant information when performing forensic accounting expertise, with the aim of identifying the necessary evidence base. In the article, special attention is paid to modern digital technologies, which allow, in the course of routine expert research, to analyze a large amount of information on economic operations reflected in accounting documents. It is emphasized that the considered approach can significantly reduce the time of expert examinations, which in its turn allows experts to properly pay attention to the analysis of data, studying documents and solving tasks of a diagnostic nature during the forensic accounting expertise. The authors come to the conclusion that an accelerated search for the necessary information from primary accounting documents, financial records, etc. makes the examination process more efficient and has a positive effect on its cost indicators. The purpose of this study is to analyze already published research regarding the benefits of software currently used in forensic accounting expertise. In the article a particular attention is paid to both the basic terms and definitions used in the current norms and rules of forensic accounting expertise, as well as to various software applications that can increase the efficiency of forensic accounting expertise. It is emphasized that to conduct forensic accounting expertise, two digital software are mainly used, which at the initial stage extract and systematize information, and at the final stage analyze the identified factorological material. The characteristic features of the applicability of financial analysis methods are highlighted and described in the article, particularly, special attention is paid to the methodological approach to the analysis of the reliability of financial records, based on the use of statistical patterns identified by F. Benford. Significant attention in the article is also paid to the development of digitalization of the economy of Kazakhstan. In this context, an attempt was made to trace the process of introducing digital tools into forensic accounting research. The authors, taking into account the analysis of various software applications used in similar expert practice, proposed methodological approaches to improve the process of conducting forensic accounting expertise. Keywords: digital tools, fraud, Benford’s Law, forensic expertise, digitalization, accounting records

Reinforcement learning based agents for improving layouts of automotive crash structures

The topology optimization of crash structures in automotive and aeronautical applications is challenging. Purely mathematical methods struggle due to the complexity of determining the sensitivities of the relevant objective functions and restrictions according to the design variables. For this reason, the Graph- and Heuristic-based Topology optimization (GHT) was developed, which controls the optimization process with rules derived from expert knowledge. In order to extend the collected expert rules, the use of reinforcement learning (RL) agents for deriving a new optimization rule is proposed in this paper. This heuristic is designed in such a way that it can be applied to many different models and load cases. An environment is introduced in which agents interact with a randomized graph to improve cells of the graph by inserting edges. The graph is derived from a structural frame model. Cells represent localized parts of the graph and delineate the areas where agents can insert edges. A newly developed shape preservation metric is presented to evaluate the performance of topology changes made by agents. This metric evaluates how much a cell has deformed by comparing its shape in the deformed and undeformed state. The training process of the agents is described and their performance is evaluated in the training environment. It is shown how the agents and the environment can be integrated as a new heuristic into the GHT. An optimization of the frame model and a vehicle rocker model with the enhanced GHT is carried out to assess its performance in practical optimizations.

Mapping Small-Scale Irrigation Areas Using Expert Decision Rules and the Random Forest Classifier in Northern Ethiopia

The mapping of small-scale irrigation areas is essential for food security and water resource management studies. The identification of small-scale irrigation areas is a challenge, but it can be overcome using expert knowledge and satellite-derived high-spatial-resolution multispectral information in conjunction with monthly normalized difference vegetation index (NDVI) time series, and additional terrain information. This paper presents a novel approach to characterize small-scale irrigation schemes that combine expert knowledge, multi-temporal NDVI time series, multispectral high-resolution satellite images, and the random forest classifier in the Zamra catchment, North Ethiopia. A fundamental element of the approach is mapping small-scale irrigation areas using expert decision rules to incorporate the available water resources. We apply expert decision rules to monthly NDVI composites from September 2020 to August 2021 along with the digital elevation model (DEM) data on the slope, drainage order, and distance maps to derive the sample set. The samples were based on the thresholds obtained by expert knowledge from field surveys. These data, along with the four spectral bands of a cloud-free Planet satellite image composite, 12 NDVI monthly composites, slope, drainage order, and distance map were used as input into a random forest classifier which was trained to classify pixels as either irrigated or non-irrigated. The results show that the analysis allows the mapping of small-scale irrigation areas with high accuracy. The classification accuracy for identifying irrigated areas showed a user accuracy ranging from 81% to 87%, along with a producer accuracy ranging from 64% to 79%. Furthermore, the classification accuracy and the kappa coefficient for the classified irrigation schemes were 80% and 0.70, respectively. As a result, these findings highlight a substantial level of agreement between the classification results and the reference data. The use of different expert knowledge-based decision rules, as a method, can be applied to extract small-scale and larger irrigation areas with similar agro-ecological characteristics.

Early Warning of Loss and Kick for Drilling Process Based on Sparse Autoencoder With Multivariate Time Series

Complicated geological environments lead to a high risk of drilling incidents. Early warning of loss and kick for the drilling process is essential to ensure process safety. On account of the nonlinear and temporal correlation of drilling parameters, an early warning method for loss and kick based on sparse autoencoder with multivariate time series is proposed. The sparse autoencoder is utilized for multivariate time series abnormality detection of the drilling process. Abnormal drilling parameter isolation is performed through contribution analysis. Reconstruction analysis and time series segmentation approaches are integrated for abnormal time series trend evaluation. The characteristic of drilling parameters under normal operation learned by the sparse autoencoder and the property of the original time series are taken into account. The final early warning result can be obtained through expert rules based on the trend evaluation result. Case studies are presented based on the data from an actual drilling project. The experiment result shows the effectiveness of the proposed method.

Discovering stop and parking behaviors of last mile delivery vehicles for urban areas based on not well conditioned GPS traces, expert knowledge and machine learning

Nowadays urban traffic is one of the most serious challenges for local authorities around the world. This challenge implies issues regarding health, energy, safety, environment and quality of life. Achieving a fair traffic outcome is a key priority for city traffic managers while maintaining city’s attractiveness for citizens and travelers and those carrying out commercial activities. As a part of this challenge, delivery and similar companies are relevant stakeholders in optimizing their routes, deliveries, and operational services. It is in this respect that urban traffic and its regulations play a key role.The conditions in which urban traffic takes place thus have to be known and one way to deal with this is based on available and accessible data. In the IoT age, there could be huge amounts of available data generated by countless sensors and systems involved in our lives, cities, infrastructures, etc.... Taking advantage of such data, when the accessibility and quality is good enough, can help us to achieve the desired goals concerning the urban traffic and its consequences. However, in practice, the availability and access to such data is currently a very serious challenge.Following on with this, data generated from the ever-increasing number of sensors on board vehicles could be very useful, not only for checking the vehicle condition, but also to gain a better of the “real-time” traffic situation or to discover traffic behaviors/patterns from the said data. In this work, a real case based study has been carried out gathering basic real GPS information regarding delivery vehicles in a city environment and OpenData to “discover” where, when and how long time delivery vehicles use the regulated parking zones for loading/unloading in the city center. Based on Expert Rules, a stop detection criteria is defined and formulated to be applied to real cases in urban areas, focusing on city centers and Machine Learning techniques to discover stop and park behaviors on last mile deliveries in a real urban area.All this is used to plan traffic strategies and facilities which can permit better and more fluent services. On the other hand, the results provide invaluable knowledge support for the expert knowledge of mobility managers, while also supplying new “findings” about the daily challenges, showing that machine learning techniques and other linked technologies are powerful tools for this challenge.

Efficient Focus Autoencoders for Fast Autonomous Flight in Intricate Wild Scenarios

The autonomous navigation of aerial robots in unknown and complex outdoor environments is a challenging problem that typically requires planners to generate collision-free trajectories based on human expert rules for fast navigation. Presently, aerial robots suffer from high latency in acquiring environmental information, which limits the control strategies that the vehicle can implement. In this study, we proposed the SAC_FAE algorithm for high-speed navigation in complex environments using deep reinforcement learning (DRL) policies. Our approach consisted of a soft actor–critic (SAC) algorithm and a focus autoencoder (FAE). Our end-to-end DRL navigation policy enabled a flying robot to efficiently accomplish navigation tasks without prior map information by relying solely on the front-end depth frames and its own pose information. The proposed algorithm outperformed existing trajectory-based optimization approaches at flight speeds exceeding 3 m/s in multiple testing environments, which demonstrates its effectiveness and efficiency.

An Intelligent Machine Learning Approach for Fraud Detection in Medical Claim Insurance: A Comprehensive Study

Medical claim insurance fraud poses a significant challenge for insurance companies and the healthcare system, leading to financial losses and reduced efficiency. In response to this issue, we present an intelligent machine- learning approach for fraud detection in medical claim insurance to enhance fraud detection accuracy and efficiency. This comprehensive study investigates the application of advanced machine learning algorithms for identifying fraudulent claims within the insurance domain. We thoroughly evaluate several candidate algorithms to select an appropriate machine learning algorithm, considering the unique characteristics of medical claim insurance data. Our chosen algorithm demonstrates superior capabilities in handling fraud detection tasks and is the foundation for our proposed intelligent approach. Our proposed approach incorporates domain knowledge and expert rules, augmenting the machine learning algorithm to address the intricacies of fraud detection within the insurance context. We introduce modifications to the algorithm, further enhancing its performance in detecting fraudulent medical claims. Through an extensive experimental setup, we evaluate the performance of our intelligent machine-learning approach. The results indicate significant accuracy, precision, recall, and F1-score improvements compared to traditional fraud detection methods. Additionally, we conduct a comparative analysis with other machine learning algorithms, affirming the superiority of our approach in this domain. The discussion section offers insights into the interpretability of the experimental findings and highlights the strengths and limitations of our approach. We conclude by emphasizing the significance of our research for the insurance industry and the potential impact on the healthcare system's efficiency and cost-effectiveness.

MACHINE LEARNING APPLICATION IN RESPONSE TO DISASTER RISK REDUCTION OF FOREST AND PEATLAND FIRE

Peat forest is a natural swamp ecosystem containing buried biomass from biomass deposits originating from past tropical swamp vegetation that has not been decomposed. Once it burns, smoldering peat fires consume huge biomass. Peat smoldering fires are challenging to extinguish. These will continuously occur for weeks to months. Experts and practitioners of peat smoldering fires are the most recommended effort to prevent them before they occur with the strategy: 'detect early, locate the fire, deliver the most appropriate technology.' Monitoring methods and early detection of forest and land fires or 'wildfire' have been highly developed and applied in Indonesia, for example, monitoring with hotspot data, FWI (Fire Weather Index), and FDRS (Fire Danger Rating System). These 'physical simulator' based methods have some weaknesses, and soon such methods will be replaced by the Machine Learning method as it is developing recently. What about the potential application of Machine Learning in the forest and land fires, particularly smoldering peat fires in Indonesia? This paper tries to answer this question. This paper recommends a conceptual design: impact-based Learning for Disaster Risk Reduction (DRR) of Forest, Land Fire, and Peat Smouldering.

Explainable Automated TI-RADS Evaluation of Thyroid Nodules.

A thyroid nodule, a common abnormal growth within the thyroid gland, is often identified through ultrasound imaging of the neck. These growths may be solid- or fluid-filled, and their treatment is influenced by factors such as size and location. The Thyroid Imaging Reporting and Data System (TI-RADS) is a classification method that categorizes thyroid nodules into risk levels based on features such as size, echogenicity, margin, shape, and calcification. It guides clinicians in deciding whether a biopsy or other further evaluation is needed. Machine learning (ML) can complement TI-RADS classification, thereby improving the detection of malignant tumors. When combined with expert rules (TI-RADS) and explanations, ML models may uncover elements that TI-RADS misses, especially when TI-RADS training data are scarce. In this paper, we present an automated system for classifying thyroid nodules according to TI-RADS and assessing malignancy effectively. We use ResNet-101 and DenseNet-201 models to classify thyroid nodules according to TI-RADS and malignancy. By analyzing the models' last layer using the Grad-CAM algorithm, we demonstrate that these models can identify risk areas and detect nodule features relevant to the TI-RADS score. By integrating Grad-CAM results with feature probability calculations, we provide a precise heat map, visualizing specific features within the nodule and potentially assisting doctors in their assessments. Our experiments show that the utilization of ResNet-101 and DenseNet-201 models, in conjunction with Grad-CAM visualization analysis, improves TI-RADS classification accuracy by up to 10%. This enhancement, achieved through iterative analysis and re-training, underscores the potential of machine learning in advancing thyroid nodule diagnosis, offering a promising direction for further exploration and clinical application.

Gaining consensus on expert rule statements for acute respiratory failure digital twin patient model in intensive care unit using a Delphi method.

Digital twin technology is a virtual depiction of a physical product and has been utilized in many fields. Digital twin patient model in healthcare is a virtual patient that provides opportunities to test the outcomes of various interventions virtually without subjecting an actual patient to possible harm. This can serve as a decision aid in the complex environment of the intensive care unit (ICU). Our objective is to develop consensus among a multidisciplinary expert panel on statements regarding respiratory pathophysiology contributing to respiratory failure in the medical ICU. We convened a panel of 34 international critical care experts. Our group modeled elements of respiratory failure pathophysiology using directed acyclic graphs (DAGs) and derived expert statements describing associated ICU clinical practices. The experts participated in three rounds of modified Delphi to gauge agreement on 78 final questions (13 statements with 6 substatements for each) using a Likert scale. A modified Delphi process achieved agreement for 62 of the final expert rule statements. Statements with the highest degree of agreement included the physiology, and management of airway obstruction decreasing alveolar ventilation and ventilation-perfusion matching. The lowest agreement statements involved the relationship between shock and hypoxemic respiratory failure due to heightened oxygen consumption and dead space. Our study proves the utility of a modified Delphi method to generate consensus to create expert rule statements for further development of a digital twin-patient model with acute respiratory failure. A substantial majority of expert rule statements used in the digital twin design align with expert knowledge of respiratory failure in critically ill patients.

The Use of Machine Learning for Image Analysis Artificial Intelligence in Clinical Microbiology.

The growing transition to digital microbiology in clinical laboratories creates the opportunity to interpret images using software. Software analysis tools can be designed to use human-curated knowledge and expert rules, but more novel artificial intelligence (AI) approaches such as machine learning (ML) are being integrated into clinical microbiology practice. These image analysis AI (IAAI) tools are beginning to penetrate routine clinical microbiology practice, and their scope and impact on routine clinical microbiology practice will continue to grow. This review separates the IAAI applications into 2 broad classification categories: (i) rare event detection/classification or (ii) score-based/categorical classification. Rare event detection can be used for screening purposes or for final identification of a microbe including microscopic detection of mycobacteria in a primary specimen, detection of bacterial colonies growing on nutrient agar, or detection of parasites in a stool preparation or blood smear. Score-based image analysis can be applied to a scoring system that classifies images in toto as its output interpretation and examples include application of the Nugent score for diagnosing bacterial vaginosis and interpretation of urine cultures. The benefits, challenges, development, and implementation strategies of IAAI tools are explored. In conclusion, IAAI is beginning to impact the routine practice of clinical microbiology, and its use can enhance the efficiency and quality of clinical microbiology practice. Although the future of IAAI is promising, currently IAAI only augments human effort and is not a replacement for human expertise.

QUIRMIA-A Phenotype-Based Algorithm for the Inference of Quinolone Resistance Mechanisms in Escherichia coli.

Quinolone resistance in Escherichia coli occurs mainly as a result of mutations in the quinolone-resistance-determining regions of gyrA and parC, which encode the drugs' primary targets. Mutational alterations affecting drug permeability or efflux as well as plasmid-based resistance mechanisms can also contribute to resistance, albeit to a lesser extent. Simplifying and generalizing complex evolutionary trajectories, low-level resistance towards fluoroquinolones arises from a single mutation in gyrA, while clinical high-level resistance is associated with two mutations in gyrA plus one mutation in parC. Both low- and high-level resistance can be detected phenotypically using nalidixic acid and fluoroquinolones such as ciprofloxacin, respectively. The aim of this study was to develop a decision tree based on disc diffusion data and to define epidemiological cut-offs to infer resistance mechanisms and to predict clinical resistance in E. coli. This diagnostic algorithm should provide a coherent genotype/phenotype classification, which separates the wildtype from any non-wildtype and further differentiates within the non-wildtype. Phenotypic susceptibility of 553 clinical E. coli isolates towards nalidixic acid, ciprofloxacin, norfloxacin and levofloxacin was determined by disc diffusion, and the genomes were sequenced. Based on epidemiological cut-offs, we developed a QUInolone Resistance Mechanisms Inference Algorithm (QUIRMIA) to infer the underlying resistance mechanisms responsible for the corresponding phenotypes, resulting in the categorization as "susceptible" (wildtype), "low-level resistance" (non-wildtype) and "high-level resistance" (non-wildtype). The congruence of phenotypes and whole genome sequencing (WGS)-derived genotypes was then assigned using QUIRMIA- and EUCAST-based AST interpretation. QUIRMIA-based inference of resistance mechanisms and sequencing data were highly congruent (542/553, 98%). In contrast, EUCAST-based classification with its binary classification into "susceptible" and "resistant" isolates failed to recognize and properly categorize low-level resistant isolates. QUIRMIA provides a coherent genotype/phenotype categorization and may be integrated in the EUCAST expert rule set, thereby enabling reliable detection of low-level resistant isolates, which may help to better predict outcome and to prevent the emergence of clinical resistance.

In vitro and in vivo susceptibility to cefalexin and amoxicillin/clavulanate in canine low-level methicillin-resistant Staphylococcus pseudintermedius.

Methicillin-resistant Staphylococcus pseudintermedius (MRSP) lineages harbouring staphylococcal cassette chromosome (SCC) mec types IV, V and ΨSCCmec57395 usually display low oxacillin MICs (0.5-2 mg/L). To evaluate how oxacillin MICs correlate with PBP mutations and susceptibility to β-lactams approved for veterinary use. Associations between MICs and PBP mutations were investigated by broth microdilution, time-kill and genome sequence analyses in 117 canine MRSP strains harbouring these SCCmec types. Clinical outcome was retrospectively evaluated in 11 MRSP-infected dogs treated with β-lactams. Low-level MRSP was defined by an oxacillin MIC <4 mg/L. Regardless of strain genotype, all low-level MRSP isolates (n = 89) were cefalexin susceptible, whereas no strains were amoxicillin/clavulanate susceptible according to clinical breakpoints. Exposure to 2× MIC of cefalexin resulted in complete killing within 8 h. High (≥4 mg/L) oxacillin MICs were associated with substitutions in native PBP2, PBP3, PBP4 and acquired PBP2a, one of which (V390M in PBP3) was statistically significant by multivariable modelling. Eight of 11 dogs responded to systemic therapy with first-generation cephalosporins (n = 4) or amoxicillin/clavulanate (n = 4) alone or with concurrent topical treatment, including 6 of 7 dogs infected with low-level MRSP. Oxacillin MIC variability in MRSP is influenced by mutations in multiple PBPs and correlates with cefalexin susceptibility. The expert rule recommending that strains with oxacillin MIC ≥0.5 mg/L are reported as resistant to all β-lactams should be reassessed based on these results, which are highly clinically relevant in light of the shortage of effective antimicrobials for systemic treatment of MRSP infections in veterinary medicine.

Artificial Situational Awareness Assessment of a Novel ATC Support System

This paper presents the application of an existing situational awareness framework to a newly developed artificial intelligence system to determine its awareness level. The system incorporates diverse automation techniques - knowledge graph, expert rules, machine learning - for gaining situational awareness and applying it in the field of air traffic control. Since the system was developed to serve as a foundation for exploring automation and artificial situational awareness, the primary result of this work is the system’s overall awareness level assessment and the identification of sub-systems that may be improved for additional awareness. The framework used was chosen in the fundamental project documents and its use proved beneficial as it enabled the demonstration of how general guidelines can be interpreted for a specific system. It also informed possible routes for improvement of the process. Highest priority awareness-related improvements are those dealing with robustness, whose implementation would substantiate the current awareness assessment. The system is shown to be on the highest awareness level conditionally, considering its proof-of-concept level. The high level reached by the system is contingent on awareness concept and condition interpretations. With the appropriate assessment of the system, implementation in an operational environment is more feasible.

Population pharmacokinetics and pharmacokinetic/pharmacodynamic evaluation of marbofloxacin against Coagulase-negative staphylococci, Staphylococcus aureus and Mycoplasma agalactiae pathogens in goats

Marbofloxacin is a broad-spectrum fluoroquinolone, and an extra-label use has been reported in horse, sheep and goat. However, extrapolation of dosage regimens from cattle to horse and small ruminants could lead to incorrect dosing due to pharmacokinetic differences among species, increasing the risk of antimicrobial resistance or toxicity. Pharmacokinetic properties of marbofloxacin, including PK/PD analysis, have been studied by intravenous, intramuscular and subcutaneous administration in lactating and non-lactating goats. A population pharmacokinetic model of marbofloxacin in goats was built using 10 pharmacokinetic studies after intravenous, intramuscular, and subcutaneous administration at a dose of 2, 5 and 10 mg/kg. Serum or plasma and milk concentration-time profiles were simultaneously fitted with a non-linear mixed effect model with Monolix software. Level of milk production (lactating and non-lactating) and health status (healthy and un-healthy) were retained as covariates on volume of distribution and clearance. Marbofloxacin concentrations were well described in plasma/serum and milk by the population model. Simulated dose regimens of marbofloxacin administered at 2, 5 and 10 mg/kg by intramuscular route for five days were evaluated (n = 5000 per group). Steady-state fAUCs for each dose regimen were obtained. Probability of target attainment of fAUC/MIC ratios were determined and PK/PDco values (highest MIC for which 90% of individuals can achieve a prior numerical value of the fAUC/MIC index) were established using Monte Carlo simulations (n = 50,000). MIC values for wild type isolates of Staphylococcus aureus, coagulase negative staphylococci, and Mycoplasma agalactiae were determined and tentative epidemiological cutoff (TECOFF) were obtained at 1.0, 0.5 and 0.5 mg/L, respectively. The PK/PDco for the dose regimen of 2 mg/kg/24 h and 5 mg/kg/24 h (0.125 and 0.25 mg/L) were lower than TECOFF (0.5 and 1 mg/L). The dosage regimen of 10 mg/kg/24 h was adequate for intermediate MIC values of 0.125–0.50 mg/L and could be effective for a population with a target fAUC/MIC ratio ˂ 48 for Coagulase negative staphylococci and Mycoplasma agalactiae, but not for Staphylococcus aureus. Results obtained in this study could be taken as a starting point by committees that set the clinical breakpoints and justifies expert rules to optimize marbofloxacin dose regimens.

A Hybrid Approach of the Deep Learning Method and Rule-Based Method for Fault Diagnosis of Sucker Rod Pumping Wells

Accurately obtaining the working status of the sucker rod pumping wells is a challenging problem for oil production. Sensors at the polished rod collect working data to form surface dynamometer cards for fault diagnosis. A prevalent method for recognizing these cards is the convolutional neural network (CNN). However, this approach has two problems: an unbalanced dataset due to varying fault frequencies and similar dynamometer card shapes that complicate recognition. This leads to a low accuracy of fault diagnosis in practice, which is unsatisfactory. Therefore, this paper proposes a hybrid approach of the deep learning method and rule-based method for fault diagnosis of sucker rod pumping wells. Specifically, when the CNN model alone fails to achieve satisfactory accuracy in the working status, historical monitoring data of the relevant wells can be collected, and expert rules can assist CNN to improve diagnostic accuracy. By analyzing time series data of factors such as the maximum and minimum loads, the area of the dynamometer card, and the load difference, a knowledgebase of expert rules can be created. When performing fault diagnosis, both the dynamometer cards and related time series data are used as inputs. The dynamometer cards are used for the CNN model to diagnose, and the related time series data are used for expert rules to diagnose. The diagnostic results and the confidence levels of the two methods are obtained and compared. When the two diagnostic results conflict, the one with higher confidence is preserved. Out of the 2066 wells and 7 fault statuses analyzed in field applications, the hybrid approach demonstrated a 21.25% increase in fault diagnosis accuracy compared with using only the CNN model. Additionally, the overall accuracy rate of the hybrid approach exceeded 95%, indicating its high effectiveness in diagnosing faults in sucker rod pumping wells.

IT-based regulation of personal health: Nudging, mobile apps and data

Mobile health applications and devices (“mobile health apps”) play increasingly important roles in the lives of individuals interested in self-regulating their personal health behaviors. While some appear to be simply consumer products and services, many are embedded in regulatory programs aimed at compliance with expert guidelines. In this paper, we draw on de Vaujany et al.’s framework for organizational IT-based regulation systems to consider how systems operate in open and distributed contexts in which actors have strong agency and regulation is indirect and voluntary. To do so, we consider how IT artifacts become embedded in practices, how data are implicated in regulatory feedback loops, and how individual, organizational and technological actors are mobilized and with what regulatory outcomes. We develop an instrumental case study as a vignette of five regulatory episodes (continuous glucose monitoring systems used by persons with diabetes) to examine how expert rules materialized in mobile health apps, data about bodily states, and IT features such as displays and alarms “nudge” individuals towards compliance with self-regulatory guidelines and practices. Through this analysis, we identify two related regulatory affordances of mobile health apps for predicting and surveilling personal health. We theorize how multilevel networks composed of trifecta of rules, IT artifacts, and practices develop as a regulatory lattice through which social regulation is realized. We conclude by considering the broader implications of this analytical approach to study voluntary, data-enriched regulatory systems.

Real-Time Integrated Energy Management Strategy Applied to Fuel Cell Hybrid Systems

Integrating hydrogen fuel cell systems (FCS) remains challenging in the expanding electric vehicle market. One of the levers to meet this challenge is the relevance of energy supervisors. This paper proposes an innovative energy management strategy (EMS) based on the integrated EMS (iEMS) concept. It uses a nested approach combining the best of the three EMS categories (optimization-based (OBS), rules-based (RBS), and learning-based (LBS) strategies) to overcome the real-time operating condition limitations of the fuel cell hybrid electric vehicle (FCHEV). Through a fuel cell/battery hybrid architecture, the purpose is to improve hydrogen consumption and manage the battery state of charge (SOC) under real-time driving conditions. The proposed iEMS approach is based on an OBS with optimal control to make the energy-optimal decision. However, it requires the adaptations of real-time operating conditions and a dynamic SOC horizon management. These requirements are supported by combining an RBS based on expert and fuzzy rules to compute the SOC target on each sliding window and an LBS based on fuzzy C-mean clustering to enhance the cooperative environment data processing and adapt it to the FHCEV topology. Our approach obtained simple and realistic system behaviors while having an acceptable computing time suitable for real time constraint. It was then designed and validated using a 27-h real-time measured database. The results show the effectiveness of the proposed iEMS concept with an excellent performance close to the optimal offline strategy (an under 2% consumption gap).

The Power of Secret Knowledge: The RAND Corporation, Ignorance Studies and Sociology

The following contribution is concerned with the relation of Dayé’s work on the RAND Corporation during the Cold War and the field of ignorance studies. In doing so, I aim to emphasise the interconnection of three central themes that pervade Dayé’s work: secrecy, ignorance, and power. In an era of Cold War insecurity, marked by strategic attempts by both sides to obscure their own capabilities, a largely secretive organization emerged as a reliable source of knowledge, helping to guide decision-making in uncertain times and to generate policy recommendations. This not only raises significant questions about the power of certain groups or individuals to define what counts as policy-guiding knowledge, it also points to a form of ignorance that is highly productive. It not only affords the creation of new knowledge practices, but it becomes a force in itself that mobilizes the creation of further ignorance. While these connections are implicit in Dayé’s work, this study seeks to bring them to the forefront and to explore them in dialogue with classical sociological literature and in the context of seminal contributions to the field of ignorance studies. In order to do so, I will start with a brief elaboration on the secrecy that surrounded the work conducted at the RAND corporation, alongside a brief discussion of the notion of secrecy and elite power in the canon of classical literature in sociology, to then introduce the field of ignorance studies. From this angle, I will explore how a particular form of ignorance lies at the core of the workings at RAND and how ignorance studies might help to better understand the developing influence and rule of experts.