Loading...
Loading...
Contributors:
Dataset URI:
10.5281/zenodo.22904586This document provides the reproducibility and audit package supporting the component-ablation analysis reported in the manuscript \emph{Ladder-of-Thought Semantic (LoT-Sem): Semantics-Governed and Traceable IEC~61131-3 Ladder Logic Synthesis and Engineering Evaluation}. Its primary purpose is to document, reproduce, and audit the one-component-at-a-time LoT-Sem ablation used for RQ3 and the corresponding manuscript ablation table. Although the complete study uses a benchmark of 25 industrial PLC use cases, the component-ablation experiment is intentionally restricted to \textbf{UC24 Industrial Robot Pick-and-Place Sequence}. UC24 is selected as a representative high-complexity case because it exercises coordinated sequence control, gripper operation, safety constraints, state transitions, and the complete LoT-Sem semantic-governance chain. Consequently, the reported ablation effects are interpreted as \emph{UC24-specific component-sensitivity evidence}, not as benchmark-wide effect estimates. The experiment compares the complete LoT-Sem-Full configuration \(\Phi_S^{\mathrm{Full}}\) with five matched one-component-at-a-time ablations that remove the explicit semantic contract, pre-code admissibility control, requirement-to-rung traceability, post-code preservation checking, or bounded local repair. All remaining experimental factors are held fixed, including the UC24 requirement, generator model and settings, LAD output contract, non-ablated prompt instructions, and method-blind external validation protocol. For each configuration, the generated Ladder artifact is frozen before independent evaluation by Gemini, Copilot, and DeepSeek. The archive preserves the exact UC24 inputs, prompt variants, frozen artifacts, validation records, aggregation data, and calculation rules used to derive the manuscript ablation results. The package therefore enables independent verification of how the reported component effects were obtained while maintaining a clear distinction between representative UC24 sensitivity analysis and benchmark-wide generalization. The generated Ladder programs remain engineering candidates and do not constitute deployment-ready or safety-certified PLC software.
Keyword(s):
MeSH Term(s):
Grants and Funding:
No Grants or Funding Data Available
Associated Publication(s):
No Associated Publication Data Available
Supplementary Data:
No Supplementary Data Available
Related Dataset(s):
No Related Dataset Data Available
The National Library of Medicine (NLM) provides information about biomedical datasets. Inclusion on an NLM Web page does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: NLM Web Policies (nih.gov)