Implementing CPSLint: A Data Validation and Sanitisation Tool for Industrial Cyber-Physical Systems
Implementing CPSLint: A Data Validation and Sanitisation Tool for Industrial Cyber-Physical Systems
Authors: Uraz Odyurt, Ömer Sayilir, Mariëlle Stoelinga, Vadim Zaytsev Date: 2026-04-20 Paper ID: openalex:2604.18191
Summary
CPSLint is a domain-specific language developed to standardize the data preparation, validation, and sanitization processes for industrial cyber-physical systems. By raising the level of abstraction, it allows users to replace repetitive, ad-hoc Python scripts with concise, maintainable code structures. The tool specifically targets the challenges of large-scale, unstructured time-series data typical of industrial monitoring. CPSLint aims to improve productivity and reproducibility for both data scientists and domain experts working with industrial data pipelines.
Key Contributions
- Introduces CPSLint, a domain-specific language (DSL) to abstract and standardize repetitive time-series data preparation, validation, and sanitization tasks.
- Reduces the complexity of data preparation in cyber-physical systems by enabling both data scientists and domain experts to define workflows concisely.
- Addresses the lack of maintainability and reproducibility inherent in ad-hoc, case-specific Python scripting for industrial data pipelines.
Open Questions & Future Work
Key Concepts
- cpslint: A domain-specific language for automating the validation and sanitization of time-series data in cyber-physical systems.
Archivist Review
I approved CPSLint as a reusable conceptual framework for DSL-based time-series data prep. I reframed the open question to be more abstract and universally applicable to industrial data DSLs, moving away from the tool-specific roadmap presented in the original submission.
Approved Concepts
- CPSLint: It introduces a domain-specific language approach to standardize and abstract repetitive data preparation tasks in industrial time-series settings, which is a novel structural intervention compared to ad-hoc scripting.
Approved Open Questions
- Industrial DSL Expressivity Bottleneck: These features address the bottleneck of limited expressiveness and format compatibility, which currently restricts the tool from being used across a wider range of industrial data-centric workflows.
Rejected Candidates
- [open_question] Extending CPSLint with domain-specific features (
cpslint-domain-specific-extensions) - other: The original title and description were overly tied to a specific tool’s roadmap; renamed to a more general research problem.
Links
Metadata & Links
- url
- https://arxiv.org/abs/2604.18191
- paper_id
- 2604.18191
- paper_source
- openalex
- domain
- time-series
- tags
-
- architectures
-
- datasets
-
- concept_slugs
- cpslint
- dataset_slugs
-
- skill
- TimeSeriesSkill
- created_at
- 2026-04-23T05:48:03Z