13780 research outputs found
Sort by
Predictive Maintenance: Leveraging Hybrid LSTM-CNN Architectures for Enhanced Performance
No full textPredictive maintenance (PdM) is an essential pillar of Industry 4.0, aiming to reduce operational downtime, extend equipment life, and enhance cost efficiency. This thesis presents an in-depth study on the development, optimization, and evaluation of hybrid learning architecture CNN-LSTM models—for time-series-based predictive maintenance tasks. We investigate the performance of both sequential and parallel hybrid models on benchmark datasets of varying complexity: NASA C-MAPSS, N-CMAPSS, and the NASA Battery dataset.
Understanding the limitations in conventional maintenance strategies and current predictive models. Motivated by the need for higher accuracy and generalization, a simulation-based framework was developed to evaluate hybrid architectures under realistic industrial conditions. CNN layers were used for localized feature extraction, while LSTM layers captured long-term temporal dependencies in degradation patterns. Both sequential and parallel CNN-LSTM architectures were developed and tested. Sequential models stack CNN and LSTM layers, whereas the parallel configuration processes data independently through each branch before merging the results. The parallel architecture notably preserved feature integrity and mitigated signal distortion between layers, leading to better learning stability.
In the initial experiments, the sequential CNN-LSTM model achieved an RMSE of 15.03 on the C-MAPSS dataset. However, validation loss began diverging from training loss early in training, indicating overfitting and instability. Transitioning to a parallel architecture provided modest but consistent improvements. On the same dataset, the RMSE was reduced to 14.75, and the training-validation curves showed better convergence. The contrast was more significant on the N-CMAPSS dataset. There, the sequential model yielded an RMSE of 26.01, while the parallel model demonstrated a sharp improvement, reaching 13.79. This reduction of nearly 50 percent highlights the importance of architectural decisions when dealing with complex, high-dimensional data. On the NASA Battery dataset, where data is less variable, optimization through Genetic Algorithms (GA) proved highly effective. The RMSE decreased from 0.35 in the initial model to 0.091 after optimization, reflecting a 73 percent improvement.
To further explore performance tuning, both Genetic Algorithm and Hyperband optimization strategies were applied. GA showed clear advantages on the battery dataset, while Hyperband performed better under tighter resource constraints. However, on the more complex C-MAPSS and N-CMAPSS datasets, these tuning strategies produced marginal gains of only 0.33 and 2.1 percent, respectively. This suggests that while tuning enhances fine-grained performance, architectural redesign yields more significant improvements in models trained on noisy or structurally varied data. Across all datasets, the parallel CNN-LSTM model demonstrated greater stability and generalization potential than its sequential counterpart, particularly when processing data with heterogeneous patterns and multiple failure modes.
This research contributes a comprehensive pipeline for predictive maintenance, integrating preprocessing, model development, and evaluation techniques. Across all experiments, the parallel CNN-LSTM consistently demonstrated better generalization than the sequential variant, particularly in handling diverse failure modes and noisy sensor data. Compared to traditional models like Random Forests and Support Vector Machines, the proposed deep learning architectures offered more accurate RUL estimation and stronger performance on unseen test data. Furthermore, the research highlights the practical constraints of implementing predictive maintenance at scale, especially the trade-off between model complexity and interpretability, and the limits of synthetic datasets in capturing real-world failure dynamics.
In conclusion, the thesis establishes that hybrid CNN-LSTM models, especially in parallel configuration, offer a robust and scalable solution for predictive maintenance. The integration of optimization techniques further refines performance on simpler datasets, while architecture remains the critical factor for success in complex environments. These findings pave the way for future research that incorporates real-time streaming data, explores more advanced architectures such as Transformers, and investigates domain adaptation methods to bridge the gap between simulated and real-world applications
CASAR Public Lecture: From Coolidge to Trump: A Century of Racial Politics in America
Get PDFDr. Gallagher\u27s talk examined how various immigration policies in the United States have been shaped by the general public\u27s attitudes regarding the race and ethnicity of those immigrating to the U.S. His talk explores the past and present social pressures that have led to both expansive and restrictive immigration policies.
Speaker: CASAR Guest speaker: Dr.Charles Gallagher ,Professor of Sociology from La Salle University, Philadelphia, USA.
Lecture 1: Date: Wednesday, Feb. 26th, 2025 Time: 7-9 pm Venue: Room 602 - Hill House building - AUC Downtown Tahrir Campus
Lecture 2: Date: Thursday, Feb. 27th, 2025 Time: 1-2 pm Venue: GAPP Executive Room P085 - Aljameel building (Beside AUC bus gate) - AUC New Cairo Campus
Learn more about CASAR and follow us:
CASAR website: https://gapp.aucegypt.edu/academics/c...: /
Facebook: casar-auc-egypt-461110727349183
Instagram: https://instagram.com/casar2003?igshi...
Linkedin: / casar-auc-egypt-7aab6031ahttps://fount.aucegypt.edu/events_and_performances/1058/thumbnail.jp
Staying Alive: Negotiating Life with the Contamination of Kitchener Drain.
Get PDFThis thesis explores how rural communities in Kafr El Sheikh governorate, Egypt, negotiate everyday life with the contamination of Kitchener Drain, the largest agricultural drain in the central Nile Delta. Drawing on ethnographic fieldwork in three villages at the tail-end end of the drain, the study explores how water contamination shapes social and ecological, social life for farmers living in the catchment of the drain. The thesis situates the farmer’s present experience within the broader history of colonial and modernist national irrigation reforms and developments highlighting how systemic neglect and uneven governance reproduce conditions of vulnerability. At the same time, the thesis looks at contaminated spaces as simultaneously being zones of abandonment and zones of freedom from structure, where creative life forms can come to be. By bringing together anthropological theories of water, contamination, and value with a Deleuzian framework of events, the research demonstrates how Kitchener Drain is both a site of precarity and of persistence, where survival entails negotiating the entanglement of life and death
Family Involvement and Innovation Performance in Lebanese Family Small and Medium-sized Enterprises
No full textMany studies examined innovation in small and medium-sized enterprises (SMEs) and different factors impacting it. However, many studies were conducted in stable regions, which leaves a wide gap in understanding how family governance impacts innovation in SMEs operating amid turbulence, as in the case of Lebanon. Lebanese SMEs are characterized by intense family involvement in operations and management, and innovation is critical for their resilience amid turbulence. Thus, it is necessary to fill this gap and examine the effect of family involvement in the ownership and management of SMEs on their innovation performance. Guided by the positivist philosophy and deductive approach, this study employs a quantitative method with a structured survey as the data collection tool. The population of the study is SMEs in Lebanon owned by families. The estimated sample size is 150 Lebanese family-owned and managed SMEs from different sectors. Cross-sectional data will be collected during May and June 2025. This study contributes empirical evidence from an underrepresented region in the family business and innovation literature and offers practical insights for SMEs navigating instability through innovation. The study advances a more context-sensitive understanding of how family governance dynamics play out in volatile markets
Rethinking Succession: The Impact of Non-Family Management on Non-Financial Performance of Family Businesses
No full textIntroduction
This empirical research study investigates the impact of non-family management succession on non-financial performance of family businesses in Nigeria. Family businesses often experience failure during generational transitions, largely due to the absence of structured succession planning and the fear of appointing non-family members to management roles when no suitable internal successor exists. Many family business owners exhibit reluctance to transfer leadership to non-family executives, primarily due to uncertainties regarding the potential impact on organisational performance. Organisational performance in family businesses is a multidimensional construct that incorporates both financial and non-financial dimensions. Financial performance indicators typically include profitability, liquidity (and solvency), market valuation, and growth (Hamann & Schiemann, 2021). On the other hand, non-financial indicators include alignment between family and business goals, continuity and succession preparedness, governance structures and controls, strategic agility, family stakeholder engagement, and socioemotional wealth (SEW), etc. While family firm owners prioritise comprehensive business performance, non-financial indicators frequently assume greater significance given their inherent complexity and challenges in quantitative assessment.
This study makes dual contributions to family business literature. First, it addresses a significant empirical gap by examining non-family succession in African family enterprises, a context that remains understudied compared to Western and Asian counterparts. Second, it advances performance measurement by employing a multidimensional framework that evaluates five critical non-financial dimensions: (1) succession preparedness and business continuity, (2) governance effectiveness (3) strategic adaptability, (4) family stakeholder alignment and (5) socioemotional wealth preservation.
Problem Statement
Extant research indicates that non-family managers can positively influence the financial performance of family firms, particularly those with below-average performance (Fang et al., 2021). However, assessing the impact of non-family managers on non-financial performance is more nuanced. Chrisman et al. (2014) highlights the challenges non-family managers face in aligning with the non-economic goals of family firms, which are often less tangible and harder to quantify. Furthermore, Zellweger et al. (2022) suggests that family managers may outperform non-family managers in achieving non-economic goals due to their intrinsic understanding of family values and objectives. These studies underscore the complexity of evaluating non-financial performance in family businesses and highlight the need for further research in this area. Given its strategic importance to the continuity of the family firm, non-financial performance is an important consideration in the choice of a successor for the family firm. However, evidence of the effect of non-family management on non-financial performance of the family firm is sparse. This study addresses this notable gap by examining the impact of non-family management succession on key dimensions of organisational performance in Nigerian family businesses.
Objective
The objective of the study is to generate insights into the role of non-family managers in shaping effective succession outcomes, thereby contributing to both theoretical advancement and practical application in the domains of entrepreneurship and family business management.
Methodology
This study employed a convergent parallel mixed methods design (Creswell & Plano Clark, 2017) to explore the effect of non-family management succession on non-financial performance of family firms in Nigeria. Data were collected through a structured survey of 113 firms and semi-structured interviews with 25 participants to ensure both breadth and depth of analysis. Participants were purposively selected based on formal registration and a minimum of five years in operation. The survey covered 24 items reflecting five non-financial performance dimensions: continuity and succession preparedness, governance structures and controls, strategic agility, family stakeholder engagement, and socioemotional wealth (SEW). Interviews provided complementary insights into these areas. The data will be analysed using suitable tools such as SPSS for the quantitative data (descriptives and regressions), while the qualitative data will thematically coded in NVivo to support triangulation.
Expected Results
The expected effect of non-family management on any of the organisational outcomes considered in this study is nuanced and could be either positive or negative, depending on contextual factors such as degree of autonomy granted to non-family managers, ownership structure, maturity stage of the firm, and the institutional environment. Where contextual factors are favourable, non-family management is expected to improve each of the components of non-financial organisational outcomes. Otherwise, non-family management may have a negative or mixed effect on organisational outcomes.
Practical Implications
Findings from this study will be of practical relevance to a many stakeholders, including family business owners, academics, non-family executives, and policymakers. Insights from the study will help business owners to navigate key succession strategies for transferring managerial responsibilities to non-family members, while also providing evidence-based findings to support the professional roles of non-family managers. Findings will also inform the development of educational curricula and regulatory frameworks that promote the sustainability of family enterprises.
Limitations and Suggestions for Future Studies
This study has two limitations. First, the sample size of 130 survey respondents and 25 interviewees is relatively small and may not adequately represent the diversity within the Nigerian family business landscape. Second, the mixed methods approach employed does not establish causal relationships neither does it provide information about the long-term effects of non-family succession. Future studies may build on this study by expanding the sample size across multiple African countries. This may also offer the opportunity to contrast this study’s findings with those in other emerging economies, such as Asia or Latin America, to discern regional patterns and divergences. In addition, longitudinal designs may be conducted to trace organisational trajectories before and after succession events to reveal long-term dynanmics
Strategic Internationalization of a Moroccan Family Business: Richbond’s Expansion Across Africa
No full textResearch Questions and Objectives
Family businesses from emerging markets are increasingly pursuing regional expansion and strategic internationalization to diversify their activities, increase resilience, and reduce dependency on domestic markets. This research centers on Richbond, a distinguished Moroccan family-owned enterprise specializing in manufacturing and consumer goods, which has systematically developed a sophisticated internationalization strategy spanning both African regional markets and global economic landscapes. The study aims to provide a comprehensive analytical exploration of the processes, strategic considerations, and organizational dynamics that support Richbond\u27s progressive business expansion. Specifically, the research seeks to address the following critical research questions:
• What motivational factors led Richbond to pursue international growth in African markets?
• How does Richbond manage its family governance model while adapting to diverse African business environments?
• What significant operational, institutional, and strategic challenges emerged during Richbond\u27s phased international expansion across different geographical and institutional landscapes
Towards a Promising Nanoformulation for the Treatment of Parkinson’s Disease
No full textParkinson’s Disease (PD) is a growing global health concern; it is the fastest-growing neurodegenerative disease and is a leading cause of disability. Moreover, currently available treatments for PD are unable to restore damaged neurons nor are they able to reduce disease progression, but rather they provide short-term relief of symptoms and at later stages, these treatments become increasingly ineffective. Accordingly, research has aimed to explore potential therapies that may help halt or slow down the disease progression. One of those potential treatments is a c-Abl inhibitor, nilotinib, however, its clinical effectiveness was limited in PD patients. One possible explanation is linked to its poor bioavailability and brain penetration capacity. In fact, reaching and maintaining the minimum required therapeutic concentration of nilotinib (i.e.,150 nM) in the brains of patients was a challenge. Accordingly, this study focused on developing nanostructured lipid carriers (NLCs) to encapsulate nilotinib, an insoluble “brick dust” drug. By conjugating glutathione (GSH) and lactoferrin (Lf) to the NLC surface via a maleimide linker, this work aimed to enhance brain delivery and therapeutic efficacy of nilotinib. It was found that the dual-ligand functionalized NLC formulation (NLC-G-Lf) demonstrated exceptional improvements in a preclinical acute-MPTP PD mouse model. Key characteristics of the NLC-G-Lf formulation included a hydrodynamic size of 260.3 nm and a zeta potential of -15.6 mV. The formulation exhibited excellent behavioral outcomes compared to nilotinib in PEG-400 vehicle; the NLC-G-Lf formulation led to the most significant improvement in motor function, with test performances rivaling those of the naive control in the pole test. Pharmacokinetic studies further reinforced these findings, revealing that NLC-G-Lf achieved the highest systemic drug exposure, with a 6h AUC of 70,777 nM.h which is approximately 3.5 times that of the nilotinib in vehicle-treated group, suggesting a significantly improved systemic bioavailability. Interestingly, formulations containing only one ligand (either GSH or Lf) resulted in lower systemic drug exposure, likely due to increased clearance. Accordingly, there seems to be a synergy between glutathione and lactoferrin playing a crucial role in the dual-ligands formula’s success. It is predicted that the unique reaction occurring between four of Lf\u27s cysteines with glutathione, leading to the conversion of the nascent Lf protein to a folded tertiary structure, seems to provide additional stability and protection of the NLCs. By using, nanotechnology and ligand-based targeting, this work highlights a promising strategy to improve the therapeutic effect of low solubility “brick dust” drugs such as nilotinib
Ways of Living and Knowing
Get PDFAfter decades of sedentarization, the Muzeina tribe in South Sinai now has new generations born into sedentary lifestyles, never experiencing nomadism like their ancestors once did. This shift has impacted their knowledge system, traditionally shaped by interactions and kinship with nature. Exposure to social media is influencing younger generations\u27 ways of living, raising the older generations’ concerns about their Bedouin identity. Despite being one of the largest tribes in South Sinai, the Muzeina are rarely represented in literature.
This ethnography explores the impact of sedentarization and contextual developments on the Muzeina\u27s knowledge system, inherited from their ancestors. Muzeina’s relationship, and sometimes kinship, with non-human actors in ecology is critical in shaping and maintaining their traditional knowledge. Data was collected through participant observation and conversations with different generations to form solid conclusions on their mobility, ways of living, and kinship with non -human actors in the ecology, revealing the loss and retention of traditional knowledge, and their reliance, or lack thereof, on modern science. Conclusions were drawn from storytelling as well, given that storytelling is a significant practice in the Muzeina\u27s interactions and narratives.
The primary conclusion is that the Muzeina continue to practice nomadism, but they perform it differently. The Muzeina maintain bonds with non-human actors, influencing their knowledge, such as the land and its influence on their use of herbal medicine. However, younger generations\u27 weakened relationship with some actors in the ecology network threatens the retention of traditional knowledge. Some knowledge has been lost by choice, as it no longer aligns with the Muzeina\u27s ethos. New material influences are starting to impact their ethos and human-nonhuman kinship, posing a further threat to their knowledge system
Commercial Pressures and Editorial Independence: Egyptian Journalists\u27 Perceptions of Advertiser Influence in Contemporary Media
Get PDFThis thesis explores the complex interplay between commercial imperatives and editorial autonomy in Egypt’s contemporary media landscape, focusing on journalists’ perceptions of advertising influence. Drawing on Cognitive Dissonance Theory (CDT) and informed by Political Economy and Institutional Theory, the study investigates how Egyptian journalists navigate ethical tensions arising from advertiser demands—both state-linked and private—within a hybrid media system marked by economic precarity, political control, and digital disruption. Through 20 in-depth interviews with journalists, alongside insights from PR professionals, media managers, and academic experts, the research reveals a pervasive erosion of the editorial-commercial boundary. Journalists report widespread self-censorship, blurred content labelling, and managerial pressure to prioritize advertiser interests, often rationalized through coping strategies such as compliance, compartmentalization, and avoidance. Digital advertising intensifies these pressures via metrics-driven content production and native advertising formats. Despite formal regulatory frameworks mandating editorial independence, enforcement remains weak, and internal codes of ethics are often inaccessible or symbolic. The findings underscore a systemic entanglement of journalism with commercial and political forces, highlighting the psychological burden on practitioners and the normalization of ethical compromise. This study contributes to global media ethics discourse by contextualizing CDT within a non-Western authoritarian-commercial media system and offers practical implications for policy reform, newsroom leadership, and journalism education
The DUBR/miR-210-5p/RETSAT Axis in Cholesterol-Associated Steatohepatitis: Clinical Signatures and Molecular Mechanisms
Get PDFMetabolic-associated fatty liver disease (MAFLD) represents an increasing global health challenge, with cholesterol-associated steatohepatitis (CASH) emerging as a distinct pathological entity warranting investigation. While CASH demonstrates unique features compared to traditional triglyceride-driven steatohepatitis, its underlying molecular mechanisms remain poorly understood. This study investigates a novel regulatory axis involving DPPA2 Upstream Binding RNA (DUBR), miR-210-5p, and retinol saturase (RETSAT). 170 patient samples revealed progressive axis dysregulation in disease states, with RETSAT emerging as a highly accurate diagnostic marker (AUC=1.000). Using novel in vitro models of MASH and CASH, we demonstrated that miR-210-5p modulation significantly affects lipid accumulation and cellular phenotype, with more pronounced effects in cholesterol-rich environments. Mechanistic studies revealed that miR-210-5p inhibition dramatically increased lipid accumulation and altered the expression of key metabolic regulators, suggesting its role as a metabolic mediator rather than the master regulator. These findings identify new molecular mechanisms in CASH pathogenesis, establish RETSAT as a potential diagnostic biomarker, and reveal promising therapeutic targets for intervention