Automated software engineering knowledge transfer: A case study on small and medium-sized software enterprises in Thailand

Shaking-table tests of external aircon units under code protocols

• Carbonic anhydrase reliably enhances carbon capture under eco-friendly conditions • Textile contactors promote liquid wicking for efficient CO 2 reactive absorption • Bifunctional reactive dyes offer a scalable enzyme immobilization approach • Enzyme catalysis shows potential to enhance ex situ mineralization • Biocatalytic textiles offer a diverse and practical platform for CO 2 mitigation Biocatalytic textiles were developed and tested as high-efficiency gas-liquid contactors for reactive CO 2 absorption using eco-friendly solvents catalyzed by carbonic anhydrase. The testing in lab to bench-scale systems with various configurations showed that biocatalytic textiles are durable and compatible with multiple different alkaline CO 2 absorption solvents across wide working concentrations, including secondary amines, carbonates, amino acids, and abundant natural water sources like pH-adjusted seawater and spring water. Biocatalytic textile contactors proved to be remarkably robust across diverse conditions, delivering similar percent CO 2 capture regardless of inlet CO 2 concentrations. By controlling gas and liquid flows, packing height and mode of enzyme delivery, single-pass CO 2 absorption efficiencies up to 95% were achieved at lab scale. Biocatalytic textiles were able to withstand repeated washing and drying, immersion and shaking in heated solvents, ambient dry storage for many months, and continuous solvent flow testing for hundreds of hours without performance reduction. Integrated bench unit testing with aqueous MDEA solvent and biocatalytic textile packing modules achieved a CO 2 adsorption rate increase of over 200% at low 1.8 L/G when compared to traditional steel structured packing. A straightforward enzyme crosslinking technology based on fiber reactive dyes developed in the course of this work makes fabrication and scale up possible using established textile manufacturing infrastructure, and a solvent composition based on seawater and wood ash extract offers potential for ex-situ mineralization of CO 2 to permanent solid carbonates for utilization or storage.

Transitioning from total to active vitamin B12 testing may influence diagnostic categorisation and costs. In this retrospective population study, a total of 181 794 vitamin B12 results from primary care requests in a UK laboratory were analysed -90 036 using total B12 assay before transitioning, and 91 758 using active B12 assay after transitioning- according to the National Institute for Health and Care Excellence (NICE)-defined thresholds. Compared with total B12, active B12 testing identified proportionately fewer patients as deficient (2.1% vs 6.5%) or indeterminate (37.0% vs 46.2%), and more as deficiency-unlikely (60.9% vs 47.3%). Although active B12 improved classification and may reduce the need for confirmatory testing by classifying fewer individuals as indeterminate, it incurred substantially higher cost, with an estimated 26-fold increase in expenditure to identify one deficient patient based on unit test cost in the NICE guidance. While active B12 may enhance diagnostic precision, its financial impact could limit broader adoption for routine testing unless test costs decrease.

• Analyzed 25,302 questions on Mocking from StackOverflow. • Applied LDA for topic modelling and pyLDAvis for topic visualizations. • Identified 30 topics, performed categorization, constructed topic hierarchy. • Analyzed category and topic-wise question trends, question types, Q&A popularity and difficulty. Mocking is a common unit testing technique that is used to simplify tests, reduce flakiness, and improve coverage by replacing real dependencies with simplified implementations. Despite its widespread use in Open Source Software (OSS) projects, there is limited understanding of how and why developers use mocks and the challenges they face. In this study, we have analyzed 25,302 questions related to Mocking on StackOverflow to identify the challenges faced by developers. We have used Latent Dirichlet Allocation (LDA) for topic modeling, identified 30 key topics, and grouped the topics into five key categories. Consequently, we analyzed the annual and relative probabilities of each category to understand the evolution of mocking-related discussions. Trend analysis reveals that categories such as Mocking Techniques and External Services have remained consistently dominant, highlighting evolving developer priorities and ongoing technical challenges. While the questions on Theoretical category declined after 2010, posts regarding Error Handling grew notably from 2009. Our findings also show an inverse relationship between a topic’s popularity and its difficulty. Popular topics like Framework Selection tend to have lower difficulty and faster resolution times, while complex topics like HTTP Requests and Responses are more likely to remain unanswered and take longer to resolve. Additionally, we evaluated questions based on the answer status- successful, ordinary, or unsuccessful, and found that topics such as Framework Selection have higher success rates, whereas tool setup and Android-related issues are more often unresolved. A classification of questions into How, Why, What , and Other revealed that over 64 % are How questions, particularly in practical domains like file access, APIs, and databases, indicating a strong need for implementation guidance. Why questions are more prevalent in error-handling contexts, reflecting conceptual challenges in debugging, while What questions are rare and mostly tied to theoretical discussions. These insights offer valuable guidance for improving developer support, tooling, and educational content in the context of mocking and unit testing.

IntroductionPostoperative delirium affects up to 60% of cardiac surgical patients. No reliable gold standard method exists for preventing delirium after cardiac surgery. An example of patient-personal monitoring is cerebral autoregulation (CA). This study aims to highlight the personal monitoring of patients' cerebral autoregulation and to determine its relationship with postoperative delirium. Additionally, it seeks to test the hypothesis that the duration of CA impairment influences the onset of postoperative delirium.MethodsA prospective study was conducted in 2021-2023. After approval of the Ethics Committee and with the patient's written consent, 104 adult patients undergoing elective coronary artery bypass graft (CABG) with cardiopulmonary bypass (CPB) surgery were enrolled. To diagnose possible delirium, all patients underwent a Confusion Assessment Method for the Intensive Care Unit (CAM-ICU) test. CA monitoring using transcranial Doppler was performed. CA status index - Mx was recorded.ResultsOur study found that 12.5% of patients were diagnosed with delirium after on-pump CABG surgery. The total duration of cerebral autoregulation impairment (TCAI) was longer in the delirium group, 4783.0 seconds versus 4204.5 seconds (p = .047), with a cut-off at 4380s. Longer cardiopulmonary bypass (CPB) leads to prolonged CA impairment (p < .001). The mean arterial pressure (MAP) during CPB was 69.67mmHg in the non-delirium group and 74.91mmHg in the delirium group (p = .001), with a cutoff at 73.669mmHg.ConclusionsCA impairment is crucial for delirium development after cardiac surgery. The duration of the TCAI event increases the risk of delirium.

Abstract Software testing is a critical component of the classical software development lifecycle, and this principle is expected to hold true for quantum software as it evolves toward large-scale production and adherence to industry standards. Developing and testing quantum software presents unique challenges due to the non-deterministic nature of quantum information, the high dimensionality of the underlying Hilbert space, complex hardware noise, and the inherent non-local properties of quantum systems. In this work, we propose a unifying theoretical framework based on probabilistic assertions, which encompasses several testing approaches—such as quantum tomography and statistical tests—for developing practical unit tests for quantum subroutines. The framework is built upon the semantic equivalence between quantum subroutines and parameterized quantum channels, as established in this work. To address the computational complexity associated with unit testing in quantum systems, we propose incorporating context-awareness into the testing process. The trade-offs between accuracy, state space coverage, and efficiency associated with the proposed theoretical framework for quantum unit testing have been demonstrated through its application to a simple three-qubit quantum subroutine that prepares a Greenberger–Horne–Zeilinger state, as well as to subroutines within a program implementing Shor’s algorithm.

We present a methodology for automated performance regression testing at the unit level, designed to integrate seamlessly into the development workflow of high-performance computing (HPC) applications. The proposed framework emphasizes the integration of performance tests with existing functional unit tests by leveraging their shared structure, thereby enabling the development of performance test suites with minimal developer effort and supporting the early detection of performance degradations. As a proof of concept, we implement and release the framework as a publicly available Julia package. It generates architecture-agnostic test suites that remain valid across heterogeneous hardware environments. To demonstrate its effectiveness, we apply it to representative HPC use cases, including memory throughput analysis in stencil-based computations and roofline modeling of a sparse linear system solver. Experiments conducted on a diverse set of CPU and GPU architectures confirm the framework’s ability to detect performance regressions consistently, validating its potential for use in continuous performance testing pipelines.

With the rapid advancement of large language models (LLMs), code translation has become a critical yet challenging task in software engineering. Existing evaluations of this task remain unreliable due to two key limitations: the absence of specialized function-level code translation benchmarks and flawed evaluation pipelines. Specifically, current pipelines lack post-processing mechanisms to extract target code from LLMs’ inherently unstable output formats, directly resulting in non-executable translations and false negatives in evaluation. To address these issues, we introduce FuncTransEval, an enhanced multilingual benchmark. It comprises 400×4 functions across four programming languages (Python, C++, Java, and JavaScript) and 12 bidirectional translation pairs, each paired with executable unit tests for fine-grained functional validation. Using FuncTransEval, we systematically assess LLMs’ true capabilities in code translation and identify their core limitations: inadequate understanding of the implicit information such as function logic, type specifications, and cross-language differences. To mitigate these shortcomings, we propose ICAM-Trans (Implicit-Context-Aware Multi-Agent Framework for function-level code translation). ICAM-Trans is a hierarchical, autonomous multi-agents architecture that explicitly uncovers and leverages these implicit contextual information during translation. It employs a Translation Orchestrate Agent (TOA) to autonomously coordinate four specialized Context Analysis Agents (CAAs), which conduct pre-translation analyses on different aspects. Insights from these analyses guide a Context-Aware Translation Agent (CTA) to generate semantically faithful target code. Experiments on FuncTransEval demonstrate that ICAM-Trans consistently outperforms strong baselines, validating its effectiveness in achieving high-fidelity and interpretable function-level code translation.

This paper proposes an engineering-oriented teaching reform for a vocational AI Machine Vision course, targeting industrial-style print defect inspection (stain and scratch) on a reproducible synthetic defect dataset. Guided by outcome-based education (OBE), we design a project-based task chain covering data specification, synthetic defect generation, preprocessing, classical machine vision baselines (thresholding, morphology, connected components, and line-structure detection), and engineering delivery of a command-line inspection tool with standardized outputs. To support formative assessment at scale and reduce subjective grading, we implement an automated assessment pipeline that verifies input/output (I/O) contract compliance, evaluates functional correctness via unit tests (pytest), measures code quality via static analysis (pylint/flake8), and validates robustness using a perturbation suite and boundary-case tests quantified by performance degradation ΔF1. Demonstration results using representative submissions show that the proposed pipeline provides consistent, transparent grading and promotes engineering deliverables and robust behavior, providing a reusable blueprint for machine vision education aligned with industrial inspection requirements.

Full-scale shaking table testing of typical units in modular high-rise concrete buildings under seismic simulation

While gender disparities in the US workforce have received substantial attention, income inequalities among subspecialty pharmacists remain understudied. This study aimed to develop a professional profile of pediatric pharmacists in the United States, construct and test gender-specific income-determination models, and identify and compare factors influencing income among male and female pediatric pharmacists. Data were collected via a national survey targeting pediatric pharmacists. The survey included human-capital, job-related, and demographic variables. Separate income-determination models were estimated between genders, using ordinary least-squares with logged annual income as the dependent variable. Key covariates included hours worked, years of experience, administrative role, salary negotiation history, and work location. A total of 285 responses were analyzed from a 29.3% response rate. Women outnumbered men 3:1. The average annual income difference between male and female pharmacists was $10,294 (6.4%, not significant); however, regression estimates showed significant differences in annual income determinants by gender (p ≤ 0.05). Work input and experience positively influenced income for both genders, but job-related covariates were significant only for women. Working more hours and having more years of experience led to bigger pay increases for men than for women. Projected annual earnings based on model estimates revealed a statistically significant gender income gap of $10,745. The study highlights a nuanced gender disparity in income among pediatric pharmacists, with income determinants functioning differently across genders. Although the average gender pay gap was not statistically significant, regression-based projections suggest potential underlying inequities. These findings call for further research and institutional dialogue to address gender-based income disparities in specialized pharmacy fields.

Food supply chain consists of various actors, such as farmers, suppliers, distributors and retailers who add a lot of difficulty to transparency, authenticity and integrity of data. The nature of traditional centralized systems is very susceptible to manipulations, loss of data, and slow response to the crisis, which reduces consumer confidence and complicates food safety recall. To solve these problems with the help of the key concepts of blockchain immutability, decentralization and cryptographic security, this paper presents A Decentralized Blockchain Provenance Framework to Food Supply Chain Traceability. ​The suggested system uses smart contracts based on Ethereum to log all essential transactions, including harvesting and packaging, ownership changes and delivery, to have a tamper-resistant account of all the critical transactions available to all stakeholders. The interaction between producers and suppliers is done through a user friendly web interface that uses React.js and Node.js and is used to connect to the off chain storage that is provided by MongoDB to store user profiles and additional data about the products. The architecture ensures end-to-end provenance tracking without having to use complex IoT hardware, rendering it very practical to use in the real world. ​ Customers obtain immediate authentication by scanning QR codes associated with blockchain documents showing authentic profiles of the origin, transportation and status modifications to develop accountability and trust. Because the system has role based access controls and hashing algorithms, it does not allow unauthorized changes and its design is better than the legacy paper based or siloed database solutions that are vulnerable to making mistakes and fraudulent activity. Decent unit, integration, and system testing ensured healthy performance in terms of authentication, registration, tracking and verification modules.

The article addresses the pressing issue of enhancing the efficiency of promoting cloud data storage services amid intense competition in the SaaS and IaaS markets. The article analyzes current trends in cloud technology development and the critical role of internet marketing as a key tool for customer acquisition and retention in IT companies. Based on a theoretical review, market dynamics analysis, and business process modeling, the study substantiates the need to integrate digital marketing tools directly into the information system of a cloud storage provider. The “Best-of-Breed” strategy is proposed and substantiated. This approach involves targeted integration of best-in-class specialized services, such as SendPulse for email campaign automation, Google Ads for contextual advertising, and Ahrefs for SEO monitoring, rather than expensive, risky all-in-one platforms like HubSpot or Salesforce Marketing Cloud. The architecture was designed, and a microservice called “Integration Module” (acting as an API Gateway) was implemented. This ensures unified, asynchronous, and secure data exchange between the company’s core information system and external marketing platforms. The microservice is built in Python using the modern asynchronous FastAPI framework, Pydantic for input data validation, SQLAlchemy with PostgreSQL for storing on-boarding events and logs, and pytest for comprehensive unit and integration testing. Emphasis is placed on automating on-boarding and customer retention: trigger-based email sequences are launched automatically upon account status changes (e.g., the start of a trial period, transition to a paid plan, approaching the end of promotional offers), significantly improving conversion rates and customer lifetime value (LTV). The full software development lifecycle is presented: requirements analysis, business process modeling using BPMN 2.0 notation, architecture design with UML component diagrams, implementation of key endpoints, testing (including negative scenarios and validation checks), containerization with Docker, and successful production deployment. The results demonstrate that this solution optimizes marketing costs, accelerates response to user behavior, enables personalized communication, and makes the information system more flexible and scalable. The developed microservice serves as a universal tool adaptable to IT companies of any size, offering subscription-based digital services. This also provides a solid foundation for future functional expansion (integration with chatbots, behavioral analytics, referral systems, etc.). The work holds both theoretical significance and high practical value for advancing digital marketing within cloud infrastructure.

Objective : to identify clinical and immunological markers of severe acute urticaria in children. Materials and methods : 61 children with severe acute urticaria were examined, clinical methods (allergoanamnesis, urticaria activity index UAS7), immunological methods (quantitative indicators of the functioning of cellular and humoral links of the innate and adaptive immune response), statistical analysis (computer environment R). Results : it was determined that 31 children had severe acute urticaria without an established cause, urticaria with allergic or pseudoallergic genesis in the form of food hypersensitivity was noted in 15 children, in the form of drug hypersensitivity in 15 children. A hereditary allergic anamnesis was found in 31 children, and a personal allergic anamnesis in 54 patients. In 32 patients, severe acute urticaria was characterized by the presence of common symptoms, in 31 children – the development of angioedema. It was found that in children with severe acute urticaria, there was a significant increase in the expression of TLR2 and TLR4 monocytes and extremely low values of TLR9 monocyte expression, the number of NK cells was reduced, there were no changes in the spontaneous test in the phagocytic unit with a decrease in the adaptive reserves of phagocytes, increased levels of lactoferrin, IFNy, IL-6, IL-17, VEGF-A, and TGF-β1 with a decrease in IL-4 content. The processes of early activation and readiness of immunocompetent cells for apoptosis are reduced, as well as the number of regulatory T-lymphocytes. In the humoral link, with an increase in the number of B-lymphocytes, there is a decrease in the levels of secretory and serum IgА, hyperproduction of IgЕ, and accumulation of circulating immune complexes. Conclusions : the association of the incidence of severe acute urticaria in children with gender, age and with various types of hypersensitivity, hereditary and personal allergic anamnesis was revealed. Dysfunctional changes in the components of the innate and adaptive immune response in children are associated with the severity of acute urticaria and suggest their leading role in the pathogenesis of the disease, the possibility of creating a model for the prognosis of the disease.

This community service program addressed post–hydro-meteorological disaster disruptions to safe water access in Rancong Village, Kuta Blang Subdistrict, Bireuen Regency, Aceh, where flooding and environmental contamination increased the risk of unsafe household water and hindered recovery of daily activities. The program aimed to restore access to safer drinking water and strengthen community capacity to operate and maintain the service through an integrated, appropriate-technology intervention. The implementation combined participatory action cycles with community co-design, student service-learning through an impact-oriented field program, and an asset-based approach to leverage local resources for sustainability. Activities were conducted in sequential stages: stakeholder coordination and joint decision-making, rapid needs and risk assessment of water sources, engineering design and installation of an integrated multi-stage emergency drinking water filtration unit, field functionality and performance testing using a before–after measurement approach, technology transfer through hands-on training, and development of standard operating procedures and an operation-and-maintenance logbook managed by the village team. The program produced key outputs: a validated baseline of local water risks and service needs; an installed filtration unit at a community-agreed location; community training and operational guidelines to ensure consistent daily use; and an agreed operation-and-maintenance mechanism to support continuity beyond the project period. Overall, the program demonstrated that pairing emergency water treatment technology with structured community governance and practical capacity building can accelerate post-disaster recovery of basic services and provide a replicable model for similar villages affected by flooding and related hazards.

The shift from monolithic applications to microservices architecture has become a cornerstone of modern software engineering, offering unprecedented scalability, resilience, and deployment flexibility. This research paper explores the comprehensive lifecycle—design, development, and testing—of a microservices-based system utilizing the ASP.NET Core framework. By applying Domain-Driven Design (DDD) principles, we architect a decentralized system utilizing API Gateways, event-driven communication, and database-per-service patterns. The paper further details the practical development phase, addressing challenges in inter-service communication and security. Finally, a multi-tiered testing strategy encompassing unit, integration, contract, and load testing is proposed and evaluated. The findings demonstrate how ASP.NET Core's lightweight, cross-platform capabilities significantly streamline the creation and maintenance of robust, enterprise-grade distributed systems.

This paper presents NAXE, an expression evaluator framework for concurrent, resilient, and temporally aware applications, particularly in IoT domains. The key contribution is the principled integration of Interval Temporal Logic as a first-class feature alongside standard arithmetic and Boolean operations. To enhance robustness, bidirectional lazy evaluation extends short-circuit semantics to yield determinate results even with indeterminate operands. A left-to-right chaining syntax and Quantified Expressions reduce cognitive load and improve accessibility, particularly for non-programmers. The framework’s asynchronous evaluation model uses callbacks over an Abstract Syntax Tree with formal operational semantics specified through EBNF. Validation combines (i) formal correctness proofs, (ii) empirical validation (over 300 unit tests), (iii) real-world deployment in a hotel IoT system and (iv) a pilot study (n = 36). This integration advances expression evaluation by supporting interval-temporal operators, determinate outcomes under indeterminate operands, quantifiers and user-oriented syntax in a single expression-evaluation design.

The mechanism of hydrogen hardening during tests of boiler tube units

Recently, Large Language Models (LLMs) have shown promising results in code generation, and several automated test generation approaches based on LLMs have been proposed. Although these approaches achieve promising performance, they suffer from two limitations. First, they lack the intrinsic understanding of the semantic intricacies and logical constructs inherent to the focal method. Second, they ignore the diversity of the generated tests and generate tests with limited code coverage. To alleviate these two limitations, in this work, we propose a novel approach named TestCTRL that optimizes LLMs for unit test generation by the Chain-of-Thought (CoT) prompt and Reinforcement Learning (RL) strategy. Specifically, we first build a new CoT dataset, containing the focal methods, corresponding unit tests, and CoT prompts. The CoT prompt includes the intention and possible test input values. Then, the CoT dataset is used to fine-tune one LLM (i.e., CodeLlama 7B) that can be seen as the policy model in RL. Meanwhile, we fine-tune another LLM (i.e., CodeGPT) as the reward model by predicting the line coverage of the focal method and its test. Moreover, we employ the Proximal Policy Optimization (PPO) algorithm to optimize the policy model and generate unit tests. We use the Defects4J benchmark to evaluate our approach from three perspectives (i.e., naturalness, validity, and code coverage). To avoid data leakage threats, we filtered out data from the CoT dataset that have the same focal method and test case names as those in the Defects4J. The experimental results demonstrate that TestCTRL outperforms state-of-the-art baselines in line and branch coverages, respectively. Besides, TestCTRL improves bug detection performance. We also investigate the reason for the proposed approach’s superiority.