Cyber-Physical Logistics Optimisation with Load-Sensitive Assessment: An Empirical Assessment of the PANDORAS Framework

Background: City freight logistics are increasingly supported by cyber-physical and digital-twin approaches, yet many frameworks are evaluated primarily using simulated data. This study empirically assesses selected PANDORAS functions using large-scale industrial logistics records. Methods: Historical delivery sequences and coordinates from 126,073 delivery operations were used to reconstruct route–day operations and calculate inter-stop geodesic distance proxies. Product-weight data from 4439 SKU records were used to construct deterministic delivery-level load proxies and a comparative load-sensitive transport-intensity indicator. A deterministic within-route spatial reordering procedure was applied while preserving route–day membership. Results: Reordering reduced the total inter-stop geodesic distance proxy from 18,420 km to 15,870 km, corresponding to a 13.84% reduction. Load-sensitive assessment additionally captured differences in inferred carried-load profiles that are not represented by distance alone. Conclusions: Industrial logistics data can support a transparent and reproducible empirical implementation of selected PANDORAS functions. The environmental indicators should be interpreted as comparative proxies rather than measured energy consumption or emissions. Further validation requires road-network data, observed loads, vehicle characteristics, and operational telemetry.