Multi-Agent AI Simulation for Evaluating Sustainability of Urban Transportation Infrastructure
Bibliographic record
Abstract
Urban transport systems in tropical cities are under increasing strain due to rapid population growth, rising mobility demand, and climate-induced stresses. These conditions create highly complex, dynamic interactions among users, vehicles, infrastructure, and environmental factors that are difficult to capture with conventional modeling approaches. Most existing transport planning frameworks rely on static or single-agent models, limiting their ability to represent real-time feedback and adaptive behavior, particularly in tropical megacities characterized by congestion, climatic volatility, and socio-economic diversity. This paper introduces a multi-agent artificial intelligence (AI) simulation framework designed as an exploratory tool for evaluating the sustainability of urban transportation infrastructure. The proposed framework integrates dynamic systems theory with multi-agent reinforcement learning to simulate interactions among heterogeneous transport agents and infrastructure components. Synthetic data reflecting typical tropical urban conditions are employed to enable controlled experimentation across multiple scenarios, including baseline, optimized infrastructure, and adaptive AI control settings. Simulation results indicate that adaptive AI scenarios outperform baseline configurations, demonstrating 25.6% higher energy efficiency, 31.4% lower congestion, and 21.8% lower emissions in the modeled environment. These outcomes illustrate the potential of the proposed framework to support comparative sustainability evaluation rather than direct real-world performance validation.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".