Université Le Havre Normandie

This paper is about state estimation of timed probabilistic discrete event systems. The main contribution is to propose general procedures for developing state estimation approaches based on artificial neural networks. It is assumed that no formal model of the system exists but a data set is available, which contains the history of the timed behaviour of the system. This dataset is exploited to develop a neural network model that uses both logical and temporal information gathered during the functioning of the system as inputs and provides the state probability vector as output. Two main approaches are proposed: (i) state estimation of timed probabilistic discrete event systems over observations: in this case the state estimate is reconstructed at the occurrence of each new observation; (ii) state estimation of timed probabilistic discrete event systems over time: in this case the state estimate is reconstructed at each clock time increment. For each approach, the paper outlines the process of data preprocessing, model building and implementation. The presented approaches pave the way for further applications of machine learning in discrete event systems.

Due to increasing pressure to implement sustainable practices, firms often exhibit reluctance and indecision, mainly driven by concerns about profit reduction. This study examines and clarifies how firms can utilize corporate social responsibility (CSR) practices by integrating additional strategies to enhance performance. Specifically, we examine how CSR can improve firm performance through the serial mediation of green entrepreneurial orientation (GEO) and differentiation strategy. Additionally, the role of green innovation as a moderator is investigated. Data from 442 small- and medium-sized enterprises were analyzed using the “structural equation modeling” technique. Grounded in a resource-based view, the results confirm a serial mediation effect of GEO and differentiation strategy between CSR and firm performance. Furthermore, green innovation positively affects both the differentiation strategy and firm performance. Surprisingly, the results show that green innovation does not moderate either (a) the influence of GEO on differentiation strategy or (b) the relationship between differentiation strategy and firm performance. The results contribute to the literature by demonstrating how the interplay between CSR, GEO, green innovation, and differentiation strategies enhances firm performance. These findings also challenge the prevailing assumption that green innovation universally moderates performance indicators, highlighting instead that its moderating effect is context-specific and not universally applicable. A key takeaway for managers is that implementing GEO and differentiation strategies enables CSR to generate substantial benefits for firms.

Dissociative recombination of the OH⁺ ion with free electrons is modeled theoretically using a recently developed approach that is based on first-principles calculations and multichannel quantum defect theory. The coupling between the incident electron and the rovibrational motion of the ion is accounted for. The cross section of the process at collision energies 10⁻⁶–1 eV and the thermally averaged rate coefficient at 10–1000 K are evaluated. The obtained anisotropic rate coefficients agree well with the data from a recent experiment carried out at the Cryogenic Storage Ring, especially when compared to previous theoretical values, which are smaller than the experimental results by about a factor of about 30.

Real applications such as supply chains, flexible industrial systems, networked control, and urban transport often exhibit a discrete event system (DESs) structure. These systems have variable states which change at discrete times. However, as these systems evolve, a variety of limits have appeared. In this paper, we consider networks of timed event graphs (NTEGs) with disturbance transitions, which are subject to generalized mutual exclusion constraints (GMECs). An algebraic method for designing control laws to guarantee these constraints is proposed. To this end, Min-Plus algebra formalisms are used to formalize the control strategy and to deduce the corresponding controllers. The calculated control laws are translated into monitor places and connected to the initial NTEGs model to prevent violation of GMECs. The developed approaches in this study are applied to manage the replenishment policy of a supply chain in the presence of disturbances.

The estimation of the volumes of contaminated soil to be treated is a crucial step in soil remediation. Numerous techniques exist for estimating the distribution of pollutants in soils, such as inverse distance weighting, kriging, Gaussian sequential simulation, and sequential indicator simulation. Unfortunately, these methods require significant computational resources to achieve precise estimations. Moreover, both kriging and Gaussian simulation require the transformation of non-normal distributions, often seen in hydrocarbon contamination, to produce accurate results. In this paper, we propose a generative neural network to generate three-dimensional maps of contaminant distributions without prior training, and to estimate the contaminated volumes. This differentiates this work from other deep learning approaches that necessitate training data. The proposed method relies on a convolutional neural network for image reconstruction and inpainting. Rather than solely depending on the concentration of chemicals determined in the laboratory, we utilize hyperspectral imaging data from soil cores to achieve a more precise depiction of soil contaminants. We assess the proposed method using a synthetic three-dimensional dataset and a real case of hydrocarbon pollution on a polluted site in France. The method demonstrates competitive performance with efficiently managed computation time, achieved through the use of a GPU accelerator. This study offers a new, practical way to improve soil pollution management using fast, data-driven techniques.

In today's rapidly transforming world, a leader's digital capabilities have become crucial for firms to achieve better performance. This study examines the influence of digital leadership (DL) capabilities on performance (both economic and environmental) with the inclusion of multiple mediators and a moderator. Specifically, three interesting but complex relationships are examined: 1) the mediation of big data analytical capabilities (BDAC) between DL capabilities and green innovation (GI), 2) the indirect effect of GI between BDAC and performance, and 3) the moderating effect of artificial intelligence (AI) change readiness on the relationship between DL capabilities and BDAC. These relationships are tested with time-lagged data from 221 small and medium-sized enterprises (SMEs). Structural equation modeling results demonstrate that BDAC significantly mediates the relationship between DL capabilities and GI. Additionally, GI significantly mediates BDAC's effect on both economic and environmental performance. Finally, AI change readiness significantly strengthens the ties between DL capabilities and BDAC. The findings extend the leadership, AI, big data, and GI literature and highlight how their interplay can be beneficial for achieving better environmental and economic performance in the digital era.

This study investigates the critical factors influencing Digital Transformation (DT) in micro, small, and medium enterprises (MSMEs) within the post‐pandemic context. Data were collected through an online survey of 341 respondents and analyzed using Partial Least Square Structural Equation Modeling (PLS‐SEM). The results indicate that Information System (IS) capacity, competitive pressure, and government support significantly contribute to DT, while perceived benefits and complexity have weaker effects. The findings also reveal that the influence of these factors varies by enterprise size: micro‐enterprises rely on external pressures and technological capabilities, small businesses depend on internal management support and Information System Capacity (ISC), and medium enterprises require further exploration of alternative drivers. This study addresses gaps in existing research by exploring DT in a unique socio‐economic setting and contributes to the literature on MSME resilience and competitiveness. Limitations include the use of convenience sampling and a focus on a single geographic context. Practical implications include recommendations for tailored policies, skill development initiatives, and improved IS infrastructure. Social implications emphasize inclusive digitalization to enhance enterprise sustainability and reduce inequality.

BACKGROUND The risk and clinical course of anal stricture observed in Crohn’s disease remains poorly known, particularly in pediatric-onset Crohn’s disease. OBJECTIVE To investigate the long-term clinical course of anal stricture in pediatric-onset CD using data from a population-based cohort. DESIGN A retrospective observational study from a prospective population-based study. SETTINGS Population-based study in Northern France. PATIENTS All patients with a diagnosis of Crohn’s disease before the age of 17 years between 1988 and 2011 within the population-based registry EPIMAD. MAIN OUTCOME MEASURES Primary outcome was the cumulative risk of anal stricture. Secondary outcomes included include risk of anal cancer, surgery, stoma and risk factors associated with anal stricture. RESULTS A total of 1,007 patients were included (median age at diagnosis, 14.5 years; IQR, 12.0-16.1), median duration of follow-up 8.8 years (IQR, 4.6-14.2)). Among them one (0,1%) had an anal stricture at diagnosis and 26 (2.6%) during follow-up. From diagnosis, the 5- and 10-years cumulative incidence of anal stricture at was 0.6% (95% CI, 0.1-1.1) and 1.4% (95% CI, 0.5-2.3), respectively. Twenty-five (n = 25/27, 93%) patients had at least one episode of anal ulceration or fistulizing perineal Crohn’s disease. In multivariable analysis, extraintestinal manifestations (HR 2.2, 95% CI, 1.0-4.8, p = 0.0270), colonic location (L2 vs L3 HR 1.2, 95% CI 0.6-2.7, p = 0.0064) and a history of fistulizing perineal Crohn’s disease (HR 9.9, 95% CI, 4.3-22.8, p < 0.0001) were significantly associated with anal stricture. After a median follow-up of 6.2 years (2.4-10.6), 11 (41%) patients required at least one anal dilatation, and healing was observed in one patient. One patient (3.7%) had an anal cancer 7 years after stricture diagnosis, and 9 (33%) patients needed a stoma. Anal stricture was significantly associated with the need of stoma (HR 5.8, 95% CI, 2.3-14.3), p = 0.0002). LIMITATIONS It has a retrospective design which makes it prone to selection bias and residual confounding. CONCLUSION Within a population-based cohort of pediatric-onset Crohn’s diease, the 10-year cumulative incidence of anal stricture was 1.4%, with associations identified with colonic disease location, and fistulizing perianal involvement. The presence of an anal stricture was linked to a fivefold increase in the likelihood of stoma formation. See Video Abstract .

Grounded in the ability-motivation-opportunity (AMO) theory and the resource-based view (RBV), this study argues that environmental transformational leadership (ETL) enhances both economic and environmental performance through green human resource management (GHRM). It also examines the moderating role of environmental knowledge in the relationship between ETL and GHRM. Additionally, this study posits that big data analytical capabilities (BDAC) strengthen the impact of GHRM on economic and environmental performance. Using a random sampling approach, we collected multi-respondent data from 355 manufacturing firms, incorporating insights from first-level managers and top executives. Structural equation modeling was employed as the analytical technique. The findings confirm that GHRM significantly mediates the effect of ETL on both economic and environmental performance. However, contrary to expectations, environmental knowledge does not moderate the ETL–GHRM relationship. In contrast, BDAC strengthens the influence of GHRM on both economic and environmental performance. This study contributes to the leadership literature by identifying the underlying mechanism (how) and boundary conditions (when) through which ETL influences firms' performance outcomes.

Additive manufacturing (AM) revolutionizes product creation with its unique layer-by-layer construction method but faces obstacles in widespread industrial use due to quality assurance and defect challenges. Integrating machine learning (ML) into AM quality control (QC) systems presents a viable solution, utilizing ML’s ability to autonomously detect patterns and extract important data, reducing the reliance on manual intervention. This study conducts an in-depth literature review to scrutinize the role of ML in augmenting QC mechanisms within extrusion-based AM processes. Our primary objective is to pinpoint ML models that excel in monitoring manufacturing activities and facilitating instantaneous defect corrections via parameter adjustments. Our analysis highlights the efficacy of convolutional neural networks (CNNs) models in defect detection, leveraging camera-based systems for an in-depth examination of printed parts. For 1-D data processing, support vector machines (SVMs) and long short-term memory (LSTM) networks have shown significant application and effectiveness. Furthermore, the study classifies various sensors and defects that can effectively benefit from ML-driven QC approaches. Our findings accentuate the essential role of ML, especially CNNs, in detecting and rectifying production flaws and also detail the synergy between different sensor technologies in creating a comprehensive monitoring framework. By integrating ML with a multisensor approach and employing real-time corrective strategies, such as dynamic parameter adjustments and the use of advanced control systems, this research underscores ML’s transformative potential in elevating AM QC. Thus, our contribution lays the groundwork for harnessing ML technologies to ensure superior quality parts production in AM, paving the way for its broader industrial adoption.

Cross sections and thermally averaged rate coefficients for the vibrational excitation and de-excitation by electron impact on the HDO molecule are computed using a theoretical approach based entirely on first principles. This approach combines scattering matrices obtained from the UK R-matrix codes for various geometries of the target molecule, three-dimensional vibrational states of HDO, and the vibrational frame transformation. The vibrational states of the molecule are evaluated by solving the Schrödinger equation numerically, without relying on the normal-mode approximation, which is known to be inaccurate for water molecules. As a result, couplings and transitions between the vibrational states of HDO are accurately accounted for. From the calculated cross sections, thermally averaged rate coefficients and their analytical fits are provided. Significant differences between the results for HDO and H2O are observed. Additionally, an uncertainty assessment of the obtained data is performed for potential use in modeling non-local thermodynamic equilibrium (non-LTE) spectra of water in various astrophysical environments.

Potential energy curves and matrix elements of radial non-adiabatic couplings of the ²Σ⁺ and ²Π states of the NeH molecule are calculated using the electronic structure package MOLPRO, in view of the study of the reactive collisions between low-energy electrons and NeH⁺.

This study investigates the gap between how hotels present their sustainability efforts (projected green image) and how guests perceive them (perceived green image). Drawing on signaling theory and using a multiple-case approach, we combined content and sentiment analysis to examine communication strategies and guest responses. The findings reveal frequent misalignments: some practices are promoted but not noticed (greenwashing risk), while others are valued by guests but undercommunicated (green hushing). Based on these patterns, we propose a two-dimensional framework that maps four communication scenarios. The concept of green cohering-where projection and perception align-emerges as the ideal state for building credibility and trust. By introducing this framework, the study contributes to green marketing literature and offers practical guidance for hospitality businesses seeking to align sustainability messaging with guest experience. Our analysis highlights the need for communication strategies that are both operationally grounded and perceptually resonant.

Background Patients with neurogastroenterology disorders like disorders of gut–brain interaction (DGBI) and gastrointestinal (GI) motility disorders often adopt restrictive diets to manage symptoms. Without professional guidance, these patients may risk developing avoidant/restrictive food intake disorder (ARFID), potentially affecting their physical and mental health. Purpose This scoping review aimed to explore the prevalence of ARFID in patients with neurogastroenterology disorders and vice versa, the direction of their association, potential risk factors, and available treatments. Methods Following PRISMA‐ScR guidelines, we searched PubMed, Web of Science, and Cochrane. Abstracts were screened for eligibility by two independent reviewers. Key Results Eighteen studies met our inclusion criteria. The prevalence of ARFID symptoms in neurogastroenterology patients ranged from 10% to 80%, while the prevalence of neurogastroenterology disorders and related GI symptoms in ARFID patients ranged from 7% to 60%. Findings on the direction of the association between eating difficulties and GI symptom occurrence were conflicting. Patients with ARFID‐neurogastroenterology disorder overlap were more likely to be female, have a lower BMI, higher anxiety and depression levels, and poorer quality of life. Two small studies evaluating treatment for this overlap suggested promising effects of cognitive behavioral therapy (CBT). Conclusions and Inferences This review highlights heterogeneity in study designs and questions the suitability of ARFID assessment tools in this context. It also underscores gaps in understanding the underlying pathophysiology and treatment approaches. Future research should prioritize validating ARFID screening tools specific to this population and standardizing study methodologies. Improved understanding of this overlap will help healthcare professionals improve management strategies and patient outcomes.

Rainfall–runoff models are widely used in water management and flood forecasting. In this study, we present a rainfall–runoff model to forecast hourly flows based on an artificial neural network (ANN). This model was developed and applied to the Iton watershed (northwestern France) to solve the problems of nonlinearity in the rainfall–runoff relationship resulting from karst and complex hydrogeological behaviors. The model design required several steps during which we were able to identify the model parameters and create the database needed to perform the flow rate forecast. This work has resulted in an ANN model able to perform an efficient prediction up to a 48 h time horizon. These results confirm that ANN models can play an important role in forecasting the nonlinear rainfall–runoff relationship encountered in many watersheds.

Background Alice in Wonderland Syndrome (AIWS) is characterized by transient distortions in visual perception—alterations in size, shape, and spatial relationships—typically described in migraine or encephalitis. Its occurrence in neurodegenerative conditions, particularly in dementia with Lewy bodies (DLB), remains exceedingly rare. Case description This article reports a case of a 68-year-old patient with dementia with Lewy bodies (DLB; limbic-early subtype) who presented with typical DLB features alongside a brief episode of misperception—reporting that his bed had “shrunk.” Neuroimaging revealed diffuse cortical atrophy with prominent bi-hippocampal and parietal lobe involvement, and hypoperfusion on HMPAO SPECT. Conclusion This is the first reported case of AIWS in a patient with DLB. We hypothesize that selective dysfunction of high-level visual networks—particularly in the right extrastriate cortex responsible for the canonical storage of object size—may lead to an agnosia of object size. This case underscores the importance of considering AIWS within the spectrum of visual disturbances in DLB. Theoretical implications These findings provide novel insights into the neurobiology of visual cognition, aligning with Husserl’s concept of the “primordial body” (Urleib) and intuition. They suggest that disruptions in the integration of visual sensory inputs and canonical object properties may critically influence the conscious reconstruction of reality.