2026-10-10T07:20:18Z https://data.mendeley.com/oai

oai:data.mendeley.com/vbhyhzj8dh.2 2026-07-27T10:34:39Z

Bîzoi, Alexandra Neurologistics: Replication Package for an Exploratory Computational Experiment on Anticipatory Ethical Knowledge in AI-Enabled Supply Chains Mendeley Data This replication package contains the complete Python code, configuration files, automated verification tests, documentation, synthetic result tables, figures, and run reports associated with the Neurologistics exploratory computational experiment. The simulation compares alternative decision architectures under controlled synthetic conditions in three AI-enabled logistics scenarios involving tensions between: - efficiency and sustainability; - fast delivery and emissions; - cost reduction and labour rights. The evaluated architectures include an efficiency-only baseline, a fixed-rule model, the complete Neurologistics architecture, and three component-ablation variants. The package includes programmed state-updating and fixed-state conditions, Monte Carlo experiments, fixed-pressure simulations, global sensitivity analysis, uncertainty intervals, ethical–efficiency frontier analysis, boundary-region analysis, and structural-stability assessment. Some internal file names and variables retain the legacy label “learning”; throughout the manuscript and interpretation documents, these refer only to programmed computational state updating, not human, organisational, or behavioural learning. All outputs are synthetically generated from explicitly defined model assumptions and parameters. The package contains no empirical human, organisational, neurological, employee, community, emissions, or operational field data. Behaviourally informed variables are abstract computational proxies and should not be interpreted as direct measurements or reproductions of neural processes. The results describe model behaviour and structural stability within the examined design space. They do not establish external validity, causal effects, behavioural realism, predictive accuracy, or organisational effectiveness. Empirical validation would require behavioural data, stakeholder assessment, and organisational field studies. The deposited materials include: - Python source code and configuration files; - automated reproducibility and verification tests; - data and parameter dictionaries; - assumptions and empirical-calibration warnings; - full 100-seed publication outputs; - generated CSV tables and figures; - configuration and completion reports; - a structural-stability report; - a SHA-256 file manifest. Recommended software environment: Python 3.10 or later. Operations Research Decision Analysis Knowledge Management Management Science Methods Artificial Intelligence Model Bîzoi, Gabriel Dataset doi:10.17632/vbhyhzj8dh.2 oai:data.mendeley.com/vbhyhzj8dh.2 Creative Commons Attribution 4.0 International http://creativecommons.org/licenses/by/4.0 https://data.mendeley.com/datasets/vbhyhzj8dh 2026-07-27T10:34:39Z