Reproducibility supplement and reference code for "Executable Evaluation Framework for the Business Value of Logistics Digital Transformation"

Reproducibility Supplement: Evaluation Framework for Assessing the Business Value of Logistics Digital Transformation Code, workflow, and verification materials for the companion paperVersion 1.0 | 25 October 2026 Purpose and Contents This supplement explains how to reproduce all quantitative results reported in Section 4 of the companion paper. It accompanies the Python program bvit_logistics_model.py (developed using NumPy and SciPy), which performs all analyses used in the study. The program includes: The core valuation model Terminal growth and robustness testing Beta-PERT Monte Carlo simulations Spearman rank correlation analysis Delta-method variance decomposition This supplement provides: Flowcharts showing the computational process A mapping of each table and figure in the paper to the function that generates it Expected outputs and acceptable tolerance ranges for verification The complete source code listing The workbook FMCG_Logistics_Case_Calculations.xlsx recalculates every number behind Tables 1–8 and Figures 2–3 with live formulas, and the first sheet is the README. What’s in it README: purpose, a map of the sheets, a quick start, color conventions, how the calculation works, and what is static and how to refresh it. Table1_Inputs: every input, with named ranges and the exchange rate. This is the only place you need to edit. Table2 and Table3_Valuation: the year-by-year model in rupiah billions, shown in US$ thousand. Table 3 also covers the funding-need, simple-payback and IRR figures from Sections 4.2 and 6.2. Table4_TermGrowth, Table5_Robustness, Table6_PERT, Table7_MC and Table8_Sensitivity: the remaining tables, each from formulas. Fig2 and Fig3: native Excel charts for the J-curve, the Monte Carlo histogram (negative-value bins in red), and the variance-share bars. Support sheets: Coef, Scen_Engine and MC_Draws, which let Excel recompute all 10,000 trials, the robustness scenarios and the sensitivity steps. The workbook UPS_ORION_Case_Calculations.xlsx uses live formulas and has a README sheet with general instructions. What it contains Inputs: every ORION figure with its unit, type (Reported, As cited, or Model) and the page in Holland et al. (2017). I re-read that article’s full text to confirm the inputs, including the 2008–2016 driver counts. Adoption: the Figure 4 data and chart. Half of drivers were using ORION around 2014.4, about 7.4 years after the first pilot. Economics: Cost against run-rate savings under both cost figures. At US$295 million the cost equals 0.74–0.98 years of savings, as in the paper. Savings as 0.82–1.09% of U.S. small-package revenue, as in the paper. Derived per-driver, per-mile and per-minute values. ORION compared with the earlier package-flow technologies. CrossCheck: the paper’s comparison with the illustrative model, giving 1.46% of revenue, 12.2% of logistics cost and 0.73% at the lower bound. ORION falls between the lower bound and the baseline. Together, these materials enable readers to verify the calculations, replicate the findings, and understand how the results were generated. Suggested citation: Sandroto, I.V., Simatupang, T.M. (2026). Reproducibility supplement and reference code for “Executable Evaluation Framework for the Business Value of Logistics Digital Transformation” (Version 1.0) [Software]. https://doi.org/10.5281/zenodo.23653464. .