Gavekal Intelligence Software (France)

Interactive constraint systems often suffer from infeasibility (no solution) due to conflicting user constraints. A common approach to recover feasibility is to eliminate the constraints that cause the conflicts in the system. This approach allows the system to provide an explanation as: “if the user is willing to drop some of their constraints, there exists a solution”. However, this form of explanation might not be very informative. A counter-factual explanation is a type of explanation that can provide a basis for the user to recover feasibility by helping them understand what changes can be applied to their existing constraints rather than removing them. We propose an efficient approach NoPropCounter-factualXplain to find counter-factual explanations for infeasible problems. We also propose a version of this algorithm which takes into account preferences called PrefnoPropCounter-factualXplain. We showcase it's usability in real world scenario using the producer/consumer constraint which is useful in problems which involve resource allocation.

Abstract UN Regulation No. 157 was the first to establish worldwide uniform provisions for the type-approval of SAE Level 3 (L3) Automated Driving Systems (ADS) called Automated Lane Keeping Systems (ALKS). To ensure the safety of the systems, detailed test provisions were introduced, which include mandatory public road tests. The present work is focusing on public road testing and safety assessment of ALKS systems and reports on the methodology, challenges and lessons learned from a proof-of-concept experimental campaign. In line with the regulatory requirements concerning the minimum duration of system verification on public roads, various traffic circumstances and mandatory traffic scenarios were all considered in carrying out the experimental campaign. Since there was no suitable ADS on the market, a vehicle equipped with an SAE L2 system was used instead. The effort was aimed at providing a practical proof of concept for the regulatory provisions and not verifying the compliance of the vehicle under test. By carefully planning test execution, the authors managed to encounter the required traffic density conditions in all needed traffic scenarios with their necessary repetition number, within four testing days. In conclusion, the experimental campaign put in place under UN R157 proved to be a feasible exercise.

Purpose Tax noncompliance is increasingly becoming a global challenge, and numerous occurrences indicate that economies in sub-Saharan Africa are the greatest hit. In Ghana, only a few people pay direct taxes. The construction sector is critical for paying taxes, but many indigenous contractors usually fail to honor their tax obligations, which ultimately affects their business success. This study examines the link between indigenous contractors’ tax compliance and their business growth (BG). Design/methodology/approach We adopt a quantitative approach and cross-sectional survey design to collect data from construction firms’ employees. We utilize descriptive statistics and hierarchical regression modeling to analyze the data. Findings We discover that indigenous contractors’ tax compliance is above average. Prospects, profit, market share and survival are identified as indicators of BG for construction firms. Furthermore, tax compliance and BG indicators were significantly positively related. However, tax compliance has a stronger effect on BG – prospects than other indicators. We also establish that contractors’ educational level can increase the effect of tax compliance on BG indicators, except that it might not be substantial. Originality/value This study provides empirical evidence that businesses, particularly contractors, with higher tax compliance are more likely to exhibit positive business outcomes and vice versa. This study has important implications for policymakers, business owners and stakeholders, as it highlights the importance of tax compliance in supporting business success. Tax administrators can rely on this study to educate indigenous taxpayers on the value of tax compliance, specifically highlighting how it can benefit their businesses.

Abstract Reasonable royalty analysis in patent litigation often involves apportioning value between the licensor and licensee based on their relative contributions to the invention’s commercialization. Comparable licenses are often considered for this purpose under the Market Approach. However, apportionment of profits determined through an Income Approach or Cost Approach has proven challenging over the years, with methodologies like the 25 percent Rule and Nash Bargaining rejected by the courts for lacking economic rigour and specificity to case facts. This article introduces the Reasonable Profit Approach, a fact-specific and economically sound methodology to address these issues. The Reasonable Profit Approach uses a three-step process: (i) determining the Apportioned Market Value generated by the implemented invention, (ii) calculating related Incremental Profit, and (iii) deriving Reasonable Profit and Royalty based on a reasonable and acceptable profit allocation to the licensee and licensor. Grounded in economic principles and consistent with case law, this approach has been successfully applied in numerous cases without admissibility challenges. By offering a robust solution, this article provides experts with a practical framework for profit apportionment in reasonable royalty analysis.

The study presents a comprehensive framework for optimizing customer retention budget by integrating clustering, classification, and mathematical optimization techniques. The study begins with the IBM Telco dataset, which is prepared through data cleansing, encoding, and scaling. In the preliminary phase, customer segmentation is performed using K-Means clustering, with ? = 3 and ? = 4 identified as optimal based on the elbow method and Silhouette score. The configurations produced three (Premium, Standard, Low) and four (Premium, Standard Plus, Standard, Low) customer segments based on purchase preferences, which served as input features for churn prediction. In the second phase, the dataset was divided into training and test sets in an 80:20 ratio, followed by data balancing using the Synthetic Minority Over-sampling Technique (SMOTE) and Edited Nearest Neighbors (ENN). Multiple classification algorithms were evaluated, including Naive Bayes (NB), Random Forest (RF), Categorical Boosting (CatBoost), Light Gradient Boosting Machine (LightGBM), Extreme Gradient Boosting (XGBoost), Gradient Boosting (GB), Support Vector Machine (SVM), Logistic Regression (LR), K-Nearest Neighbors (KNN), and Multi-Layer Perceptron (MLP) using F1-score as the performance metric. CatBoost and LightGBM, with k values of 3 and 4, respectively, were the highest-performing classification models, with only minimal differences in performance. Ultimately, customer segmentation established customer prioritization, whereas churn prediction assessed customer churn likelihood. Four distinct configurations were assessed utilizing mixed-integer linear programming (MILP) to optimise retention budget allocation within uniform budget constraints, discount amounts, and churn thresholds. In both the k=3 and k=4 scenarios, CatBoost surpassed LightGBM, with CatBoost at K=3 effectively discounting 66% of at-risk consumers across all three segments, hence improving the intervention's efficacy and budget allocation, making it the ideal choice for maximizing customer retention. The results demonstrate the importance of segmentation in enhancing retention budgeting and budget optimization, particularly concerning parameter sensitivity.

This paper presents a preliminary study on integrating Model Context Protocol (MCP) into Computer-Aided Engineering (CAE) workflows. We developed a proof-of-concept system that combines FreeCAD for 3D modeling, Gmsh for mesh generation, and ADVENTURE for finite element analysis, orchestrated through MCP-enabled AI. The system demonstrates automated design iteration, mesh quality optimization, and boundary condition generation through natural language interactions. Results show that MCP can effectively bridge the gap between AI reasoning capabilities and specialized engineering tools, enabling more intuitive and efficient CAE workflows. Performance evaluation indicates successful automation of the complete design-to-analysis pipeline with minimal human intervention.

1.AbstractBackground and objectivesDiagnosis of non-small cell lung cancer (NSCLC) currently relies on imaging and in-clinic visits, however these methods are not effective at detecting early-stage disease. The investigation into blood-based biomarkers aims to simplify the diagnostic process and has the potential for identifying disease-associated changes before they can be seen using imaging techniques.Methods and designIn this study, plasma and frozen whole blood cell pellets from patients with NSCLC and healthy controls were processed using both classical as well as novel techniques to produce a unique set of 4 sample types from a single blood draw. Samples were analysed using 12 commercially available immunoassay kits in addition to liquid chromatography mass spectrometry using a Q Exactive HF-X Orbitrap to collectively screen 3974 proteins as potential biomarkers.Results and conclusionsAnalysis of all sample types produced a set of 522 differentially expressed proteins, with conventional blood analysis (proteomic analysis of plasma) accounting for only 7 of that total. Boosted regression tree analysis of the differentially expressed proteins produced a panel of 13 proteins that were able to discriminate between controls and NSCLC patients with an area under the ROC curve (AUC) of 0.864 for the set. Our rapid and reproducible blood preparation and analysis methods enable the production of high-quality data from small aliquots of complex samples that are typically seen as requiring significant fractionation prior to proteomic analysis.

Objetivo Los indicadores de salud pública, como el índice de mortalidad por enfermedad cardiovascular, son una herramienta para evaluar la magnitud de los problemas de salud de la población. Por ende, es necesario generar estrategias que permitan predecir su comportamiento para evaluar el cumplimiento de los objetivos y programas en salud pública. En este contexto, se propuso aplicar una metodología basada en lacaminata al azar probabilista y una ecuación de segundo grado para predecir el índice de mortalidad cardiovascular en el departamento del Magdalena. Métodos Se tomaron valores de los índices de mortalidad por enfermedad cardiovascular del departamento del Magdalena a partir de los registros históricos reportados por el Instituto Nacional de Salud entre 1998 y 2011. Posteriormente, se determinó que el comportamiento de esta variable fuera compatible con la caminata al azar para establecer espacios de probabilidad y una ecuación de segundo grado con la se predijo el valor de este índice para el 2011 con el propósito de verificar la eficacia del método. Resultados El resultado predicho de la variable fue de 93,2. Al compararlo con el valor real reportado de 91,1, se observa que el método presenta una precisión predictiva del 98%. Conclusiones Es posible predecir con alta precisión el comportamiento de indicadores en salud pública utilizando métodos fisicomatemáticos. Esto resulta de utilidad para las entidades de vigilancia epidemiológica, ya que permite evaluar la eficacia de las intervenciones poblacionales en salud pública.

Introduction: part of the effectiveness of follow-ups of patients with HIV in antiretroviral therapy is done through the quantification of CD4 + lymphocytes, hence the correct establishment of these values is an issue of interest in the clinical setting. Objective: to establish predictive mathematical relationships between CD4+ cell counts in ranges >500, [200,500], <200, between 200 and >500 and <200 up to 500 cells/μL3 with the absolute leukocyte count of patient samples over time in the context of the theory of probability. Methods: Through an inductive process carried out in 11 patient samples, mathematical patterns that forecast in time the correspondence between absolute leukocyte counts and CD4+ counts that can occur in five ranges of clinical interest. Then, a confirmation was done with 139 patients in a blind study obtaining the probability values for each range as well as sensitivity and specificity. Results: The five dynamics predicted achieved probabilities that varied between 0.96 and 1, with a global probability of 0.99 with sensitivity and specificity values of 99%. Conclusions: a self-organized mathematical temporal order that allows to forecast the values of CD4+ cells in relation to leukocyte counts in ranges of clinical interests was found, which could be useful to develop surveillance programs of HIV-infected patients in low-income countries, improving their survival rates.