Cost–Accuracy Trade-offs in Unpaved Rural Road Condition Assessment: Visual Indices and Smartphone-Based IRI in the Ecuadorian Andes

Cost-Accuracy Trade-offs in Unpaved Rural Road Condition Assessment: Visual Indices and Smartphone-Based IRI in the Ecuadorian Andes Supplementary material. Las Chinchas-Zambi corridor, Loja, Ecuador.Vasquez-Monteros, Cordova Faggioni, Ariza Flores, Alonso-Solorzano & Morea. CONTENTS--------Consolidado_resultados.xlsx Field database. Seven worksheets: four visual inspection forms (F_URCI, F_MTC II, F_PASER, F_DNV(ICPN)), the profilometer runs (IRI_ROUGHMETER), the smartphone series (IRI_APP) and the harmonized set (RESULTS). unpaved_road_analysis.py Statistical analysis. Reads the workbook and reproduces Tables 2 to 9 and A1 to A5, and Figures 4 to 10, A1 and A2, in a single execution (Table 1 compares the four methods from their manuals, Table 10 comes from the cost script, and Figures 1 to 3 are a digital elevation model and field photographs): repeatability and its confidence intervals, smartphone structure and screening capacity, regression models with leave-one-out and leave-one-block-out spatial cross-validation, HAC standard errors, moving block bootstrap, native-scale robustness and the Williams test. Writes ./outputs/statistical_results.xlsx (26 sheets) and ten figures at 300 dpi.Requires: pandas, numpy, scipy, statsmodels, matplotlib, openpyxl. Run time: about five minutes on a laptop. The cost is dominated by the resampling: 2000 replicates for every bootstrap interval and the leave-one-block-out validation at four block lengths. unpaved_road_costs.py Unit cost estimation. Deliberately separate and self-contained: it reads no data and imports nothing beyond the standard library. Every input is a parameter declared at the top of the file, with its source. Reproduces Table 10 of the paper and its sensitivity analysis. To transfer the analysis to another context, edit the block headed LOCAL CONTEXT (wages, vehicle rate, equipment price) and run the file. Requires: nothing beyond Python 3. statistical_results.xlsx Output of unpaved_road_analysis.py, provided so that the tables of the paper can be checked without running the script. figures/ The ten figures produced by the analysis script, as published (Figures 4 to 10, A1, A2, and the supplementary Figure S1). HOW TO REPRODUCE---------------- python unpaved_road_analysis.py python unpaved_road_costs.py Both scripts locate their inputs relative to their own directory, so no path needs editing. The data file is accepted with either an underscore or a space in its name. The analysis script honours the environment variables UNPAVED_DATA_FILE and UNPAVED_OUT_DIR if a different location is preferred. NOTES-----Rounding: all monetary figures are rounded to the cent and the rounded values are carried forward, so that each column of the cost table adds up to its own printed total. Resampling: seed 20260721, 2000 replicates. The bootstrap intervals are reproducible. Not reproduced by these scripts: the Passing-Bablok regression (Real Statistics) and the selection of functional forms (Statgraphics), as stated in the paper. The results reported in the paper were produced with Python 3.11.4, pandas 3.0.2, numpy 2.4.4, scipy 1.17.1, statsmodels 0.14.6, matplotlib 3.10.8 and openpyxl 3.1.5, and were reproduced without numerical differences on a second stack (Python 3.13.9 with pandas 2.3.3, numpy 2.3.5, scipy 1.16.3 and matplotlib 3.10.6). Citation: Manuscript under review