This study presents RetenNet, a comprehensive framework for managing customer churn in telecommunications, integrating predictive modelling, prescriptive optimization, and explainable artificial intelligence (XAI) incorporated with Large Language Models (LLMs). The process commences with the IBM Telco dataset, divided in an 80:20 ratio into training and testing sets. Categorical variables are converted by one-hot and label encoding, whilst class imbalance is mitigated using SMOTEENN. Min-max scaling and mutual information-based feature selection guarantee data appropriateness for machine learning models. Five classification algorithms, i.e., Random Forest (RF), Support Vector Machine (SVM), Extreme Gradient Boosting (XGB), Logistic Regression (LR), and Multi-Layer Perceptron (MLP) are assessed. The SVM model utilizing an RBF kernel exhibits optimal performance. In conjunction with nested cross-validation, Bayesian optimization guarantees excellent hyperparameter optimization and generalization. Performance is evaluated using the F1-score to highlight the implications of false negatives and false positives in churn situations. The methodology additionally incorporates fuzzy rule-based clustering, facilitating flexibility in customer segment identification for intervention priority. Prescriptive optimization uses linear integer programming to distribute retention budget according to model results and business constraints. SHAP waterfall plot employed to guarantee transparency and facilitate actionable insights. Furthermore, Gemini 1.5 Flash, a multimodal LLM, generates analysis and produces contextual recommendations derived from the SHAP waterfall plot. RetenNet offers a comprehensive and interpretable approach to the churn management pipeline, including classical machine learning, prescriptive optimization, and LLM-driven explainable artificial intelligence to enhance decision-making in customer retention efforts.

Permintaan material tubing di sektor minyak dan gas adalah elemen penting yang secara langsung berpengaruh pada kelancaran proses produksi. Tubing berfungsi sebagai saluran utama untuk mengalirkan fluida dari reservoir ke permukaan, sehingga keterlambatan dalam mendapatkan komponen ini bisa mengganggu proses pengeboran dan produksi. Salah satu kendala utama dalam pengelolaan rantai pasok adalah ketidakpastian dalam fluktuasi permintaan material, terutama dalam kondisi operasional yang berubah-ubah seperti yang dialami oleh PT Pertamina EP Asset-1 Field Jambi. Oleh karena itu, diperlukan metode peramalan yang dapat secara tepat memperkirakan permintaan di masa mendatang untuk mendukung keputusan dalam pengadaan dan pengelolaan stok. Studi ini bertujuan untuk menganalisis dan meramalkan permintaan material tubing dengan menggunakan metode Single Exponential Smoothing (SES), yang dikenal efektif untuk data yang tidak memiliki tren dan tidak musiman. Data yang dianalisis adalah permintaan tubing dari tahun 2015 hingga 2023. Dua nilai smoothing (dan ) diuji untuk menilai akurasi model dengan menggunakan indikator Mean Absolute Deviation (MAD), Mean Squared Error (MSE), dan Mean Absolute Percentage Error (MAPE). Hasil penelitian menunjukkan bahwa nilai  menghasilkan tingkat akurasi yang superior dengan MAD sebesar 479,55, MSE sebesar 614. 854,49 dan MAPE sebesar 16,8%. Temuan tersebut menggambarkan bahwa metode SES dapat berfungsi sebagai alat yang efektif untuk mendukung efisiensi pengadaan material, pengendalian persediaan, serta pengambilan keputusan strategis dalam sistem logistik perusahaan.

Keselamatan dan kesehatan kerja (K3) merupakan aspek penting dalam industri manufaktur, khususnya pada proses yang melibatkan mesin berisiko tinggi seperti spot welding. Penelitian ini dilakukan untuk menganalisis penyebab kecelakaan kerja yang terjadi pada mesin spot welding di salah satu perusahaan manufaktur, serta merancang sistem kerja yang lebih aman dengan pendekatan Kaizen. Metode yang digunakan dalam penelitian ini adalah studi kasus dengan pendekatan deskriptif-analitis. Pengumpulan data dilakukan melalui observasi langsung, studi dokumen, dan wawancara. Analisis risiko dilakukan menggunakan metode HIRARC (Hazard Identification, Risk Assessment, and Risk Control), dan akar masalah diidentifikasi menggunakan diagram fishbone. Hasil penelitian menunjukkan bahwa kecelakaan kerja terjadi akibat ketidakpatuhan operator terhadap prosedur kerja, tidak adanya sistem pengaman otomatis, serta lemahnya pengawasan. Sebagai solusi, dirancang perbaikan sistem kerja dengan menerapkan double-hand operation system untuk mencegah tangan operator berada di area berbahaya saat mesin beroperasi. Modifikasi ini dinilai efektif karena merupakan bentuk poka-yoke yang sederhana namun mampu menghilangkan risiko secara teknis. Penerapan sistem kerja berbasis Kaizen terbukti dapat meningkatkan keselamatan kerja secara signifikan dan mencegah terulangnya kecelakaan serupa di masa mendatang.

This study aims to analyze the development strategy of shoe washing and Medium Enterprises (SMEs) in facing increasingly tight competition. Shoe washing SMEs, as one of the business sectors with great potential, often face serious challenges, such as limited capital, low use of technology, and inconsistent service quality. This study uses a qualitative approach with a case study method on several urban laundry SMEs, which are analyzed using the Swot theoretical framework and Porter's Five Forces model. The results of the study show that laundry SMEs have strengths in competitive prices, proximity to customers, and strategic locations. However, the main weakness found is the lack of utilization of technology and innovation in services, which hinders business competitiveness. Existing opportunities, such as increasing awareness of sustainability-based services and digitalization, provide space for business development. The threat of competition with laundry franchises and changes in customer preferences that are increasingly moving towards fast and efficient services, are serious challenges for SMEs. Based on the results of the analysis, the recommended development strategies include the use of digital technology for management and marketing, improving service quality, innovation in types of services, and product diversification to attract new customers.

In the digital era, marketing strategies through e-commerce have rapidly evolved. Shopee, as one of the leading platforms, utilizes marketing content and live streaming features as efforts to attract customers. This study aims to analyze the influence of marketing content and Shopee live streaming on customer purchasing decisions. The research employs a descriptive quantitative approach using a questionnaire as the instrument. The expected results indicate that engaging marketing content and interactive live streaming have a positive impact on consumer interest and purchasing decisions. These findings are expected to serve as a reference for businesses in optimizing their digital marketing strategies.

This research examines the implementation of social welfare through social assistance funds in Indonesia during the Covid-19 pandemic within the framework of the welfare state theory. Using a normative juridical approach with a descriptive-normative character, the study analyzes the legal foundations, the theory of justice, and utilitarianism as conceptual bases. Two approaches Statute Approach and Conceptual Approach are employed to review key regulations (Law No. 13/2011, Law No. 11/2009, Government Regulation No. 39/2012, and other implementing regulations) as well as the concepts of distributive justice, utility, and the capability principle. Secondary data consists of primary, secondary, and tertiary legal materials collected through literature study and qualitatively analyzed. The findings reveal that programs such as Direct Cash Assistance (BLT), the Family Hope Program (PKH), and the Staple Food Card (Kartu Sembako) represent state interventions aimed at correcting inequality and ensuring citizens’ social rights. The application of the principles of social justice, utility, transparancy, accountability, participation, and sustainability has strengthened the legitimacy of social assistance as a redistributive instrument. However, the main challenges lie in the accuracy of the Integrated Social Welfare Data (DTKS), overlapping recipients, limited public participation, and long-term planning that remains reactive to crises. Evaluation of alignment with welfare state principles reveals that social assistance interventions have helped reduce short-term economic burdens but are not yet optimal in empowering recipients toward self-reliance. Based on these findings, it is recommended to regularly update the DTKS, enhance digitalization and data transparancy, expand participatory mechanisms, and integrate social assistance policies into long-term national development strategies. Thus, social assistance funds can serve not only as emergency aid but also as sustainable instruments to realize social justice and inclusive growth.

This research examines the legal responsibility of corporate guarantors declared bankrupt for their obligations to creditors in loan agreements, analyzing Court Decision No. 318/Pdt.Sus-PKPU/2022/PN Commercial Court Jakarta Central. The study addresses the legal complexities arising when a corporate guarantee becomes insolvent before the principal debtor defaults, creating jurisdictional conflicts between bankruptcy law and guarantee obligations. Using normative legal research methodology with a juridical normative approach, this study analyzes the application of Article 229(2) in conjunction with Article 278(6) of Law No. 37/2004 on Bankruptcy and Suspension of Debt Payment Obligations (K-PKPU) and Article 1381 of the Civil Code. The research reveals that the curator's action in obstructing and rejecting PT KawanCicil Teknologi Utama's claim registration against the bankrupt guarantor without proper judicial determination constitutes conduct not based on applicable laws and regulations. The findings demonstrate that when a guarantor is declared bankrupt and the debtor is in default, the Supervisory Judge and Curator must accept PT KawanCicil Teknologi Utama as a creditor with rights to the debtor's assets during asset liquidation proceedings. The study concludes that bankruptcy declaration of a guarantor does not automatically terminate guarantee obligations under Article 1381 of the Civil Code, as bankruptcy is not enumerated among the causes of contract termination. This research contributes to legal certainty in corporate guarantee enforcement within Indonesia's bankruptcy framework and provides recommendations for legislative harmonization between conflicting provisions in bankruptcy law.

The digital era has compelled MSME actors to adapt, not only in marketing and operational aspects but also in financial management. This study aims to analyze the effect of financial management on the performance of MSMEs in the digital era. The research was conducted using a quantitative approach through a survey of 120 MSME actors in West Nusa Tenggara Province who have utilized digital applications for financial record-keeping. The data were analyzed using validity and reliability tests, multiple linear regression, as well as partial and simultaneous hypothesis testing. The results indicate that sound financial management significantly affects MSME performance, with record-keeping and budget planning being the most dominant indicators.

This community service program aims to educate village residents on the importance of waste separation and processing, implement simple circular economy-based technologies, and establish a community-based waste management system. A participatory approach is carried out through awareness campaigns and training on household-level waste sorting, environmental campaigns through schools, communities, and social media, as well as incentive programs for residents actively involved in waste management, such as waste bank initiatives. The outcomes of this program are expected to create a cleaner and healthier village environment, improve the community’s economic well-being, and support the achievement of the Sustainable Development Goals (SDGs). Furthermore, the program is intended to serve as a replicable model for waste management in other regions with similar characteristics.

The complexity of financial decision-making has intensified in the digital era due to data saturation, market volatility, and the inability of conventional models to respond to real-time and non-linear dynamics. Addressing these challenges requires the integration of intelligent systems capable of adapting to evolving financial environments. Smart finance, which combines artificial intelligence, machine learning, big data analytics, blockchain, and automation, offers transformative potential across financial services. This study synthesizes scholarly findings from 2019 to 2024 across five domains: AI-based modeling, robo-advisory applications, behavioral finance integration, decentralized finance (DeFi), and real-time risk analytics. Results indicate substantial gains in efficiency, accuracy, and personalization, yet also reveal persistent challenges, including algorithmic opacity, ethical concerns in data use, and regulatory ambiguity. Advancing smart finance demands development in explainable AI, hybrid advisory systems, and inclusive, adaptive regulation for decentralized infrastructures. The scope of the analysis is limited to peer-reviewed academic literature published in English, excluding industry reports and grey literature.