Daytime vehicle lighting: a critical review of energy and emissions evidence transferability
Description
This dataset supports a critical methodological review of evidence transferability in daytime vehicle lighting. It provides transcribed published-source data, reproducible Python code, calculated results and documentation for assessing the relationship between lighting demand and vehicle energy consumption. The package contains 30 rounded fuel-consumption values for ten historical vehicle model classes from TRL report PPR170, covering baseline conditions and additional electrical loads of 42 W and 160 W, together with 20 published comparison percentages. Three Python scripts reconstruct fuel-consumption contrasts, check published driver-efficiency arithmetic and calculate display-rounding bounds for ratios sharing a common baseline. Calculations distinguish reductions in the lighting-related fuel increment from reductions in total vehicle fuel consumption. Supporting files include a data dictionary, source references and extraction locations, search and access records, an inventory of 13 publications, and seven source–target transferability assessments. A separate record documents the published electrical-prediction transfer case from Mueller et al. (2024), including target-specific adjustment and reported performance indices. All numerical research inputs originate from cited publications. The package contains no new experimental measurements or synthetic observations. It does not rerun the original vehicle models, provide contemporary fleet estimates or directly quantify emission reductions. Historical model-class results should be interpreted within their documented operating conditions and system boundaries. The scripts use the Python 3 standard library and require no network connection. Stored outputs and a file-integrity manifest are included. The README provides execution instructions, interpretation limits and information on AI assistance and author-verification status. Third-party full-text publications are not redistributed.