面向以数据和AI为中心的系统重塑软件开发生命周期（SDLC）
Reshaping the SDLC for Data- and AI-Centric Systems
浏览论文内容
中文总结 AI 辅助
本文整合DataOps、MLOps、LLMOps，提出分阶段转型描述、轻量级形式化方法、五层框架及概念研究模型，为以数据和AI为中心的系统重塑SDLC并规划研究议程。
中文摘要 AI 辅助
传统软件开发生命周期（SDLC）假设系统行为主要由源代码决定，允许通过以代码为中心的实践来指定、实现和验证正确性。数据密集型和AI驱动的系统对这一假设提出了挑战，因为它们的行为来自代码、数据和学习模型的交互，且当现实世界条件偏离训练数据时，性能可能会下降。本文研究了通过DataOps、MLOps和LLMOps将数据工程和软件工程实践整合，如何重塑这类系统的SDLC。我们做出四项贡献：第一，我们综合软件工程、数据管理、机器学习系统和以人为本计算领域的文献，形成了一个分阶段的生命周期转型描述，涵盖需求、架构、开发、测试、部署、监控、治理和组织；第二，我们提供了一种轻量级形式化方法，其中系统行为由代码、数据和模型配置定义，需求成为以评估为导向、具有概率接受域的规范，且通过基于统计的验证门控控制发布；第三，我们开发了一个自适应五层生命周期框架，包括构件层、契约层、门控层、控制层和治理层，将维护定位为配置漂移下的闭环控制问题；第四，我们提出了一个概念性研究模型，将数据工程整合与可测量的生命周期结果关联，并批判性评估了证据基础。尽管转型方向日益明确，但其规模仍未得到充分量化。我们最后提出了以经验为基础的、面向以数据和AI为中心的系统的自适应SDLC的研究议程。
英文摘要
The traditional Software Development Lifecycle (SDLC) assumes that system behavior is determined primarily by source code, allowing correctness to be specified, implemented, and verified through code-centric practices. Data-intensive and AI-enabled systems challenge this assumption because their behavior emerges from the interaction of code, data, and learned models, while performance may degrade as real-world conditions drift from training data. This paper examines how integrating data engineering and software engineering practices, operationalized through DataOps, MLOps, and LLMOps, reshapes the SDLC for these systems. We make four contributions. First, we synthesize literature across software engineering, data management, machine learning systems, and human-centered computing into a phase-structured account of lifecycle transformation spanning requirements, architecture, development, testing, deployment, monitoring, governance, and organization. Second, we provide a lightweight formalization in which system behavior is defined over code, data, and model configurations; requirements become evaluation-led specifications with probabilistic acceptance regions; and promotion is controlled through statistically grounded validation gates. Third, we develop an adaptive five-layer lifecycle framework comprising artifact, contract, gate, control, and governance layers, positioning maintenance as a closed-loop control problem under configuration drift. Fourth, we propose a conceptual research model linking data engineering integration to measurable lifecycle outcomes and critically assess the evidence base. While the direction of transformation is increasingly established, its magnitude remains insufficiently quantified. We conclude with a research agenda for an empirically grounded, adaptive SDLC for data- and AI-centric systems.
