Volume-11, Issue-10, October 2025

Abstract

Retrieval-Augmented Generation (RAG) has emerged as a powerful paradigm for enhancing the performance of Large Language Models (LLMs) by grounding their outputs on external knowledge. This paper presents a domain-specific RAG pipeline integrating Ollama with LangChain, FAISS, and Hugging Face embeddings to process and query a custom corpus of 100 research papers. By leveraging FAISS for efficient similarity search and Hugging Face models for semantic embeddings, the system enables precise and context-aware retrieval of academic knowledge. The results demonstrate improved accuracy, contextual relevance, and reduced hallucinations compared to traditional LLM usage, making the framework suitable for research assistance and literature review automation.

Keywords: Retrieval-Augmented Generation, Ollama, LangChain, FAISS, Hugging Face, Research Assistance, Large Language Model

References

Keywords: Retrieval-Augmented Generation, Ollama, LangChain, FAISS, Hugging Face, Research Assistance, Large Language Model

Abstract

Shape memory alloys (SMAs) represent a rapidly expanding class of smart materials with global market projections reaching $20.8 billion by 2030. This comprehensive review systematically compares the performance characteristics, application domains, and implementation challenges of major SMA systems including NiTi-based, Cu-based, Fe-based, and emerging high-entropy alloys. Through systematic literature analysis of studies published between 2000-2025, this work evaluates alloy families across seven key dimensions: transformation temperatures, shape memory recovery, mechanical properties, economic factors, application suitability, manufacturing readiness, and research maturity. Binary NiTi maintains superiority in functional properties with 8-10% recoverable strain and >95% shape recovery, establishing dominance in biomedical applications. NiTi-Hf variants enable high-temperature aerospace applications above 200°C, while Cu-based systems offer cost-effective alternatives with optimized compositions approaching NiTi performance levels. Fe-based systems demonstrate growing potential in civil infrastructure through superior cost-effectiveness and environmental durability. The analysis reveals critical trade-offs between performance, cost, and processability, with no single system excelling universally. Application-specific optimization proves more critical than absolute performance metrics, as different engineering domains prioritize distinct material characteristics. Manufacturing maturity varies significantly, with NiTi benefiting from extensive processing infrastructure while emerging systems require specialized development. Key research gaps include standardized testing protocols, predictive processing-property models, environmental degradation mechanisms, and lifecycle sustainability assessments. This systematic framework provides objective criteria for material selection and identifies strategic directions for advancing SMA technology adoption across diverse engineering applications.

Keywords: Shape memory alloys, superelasticity, engineering applications, material selection, smart materials

References

Keywords: Shape memory alloys, superelasticity, engineering applications, material selection, smart materials