Optimizing Maintenance Strategies for Transportation Networks: Integrating Risk Attitudes in Decision Making
Bibliographic record
Abstract
Optimizing maintenance strategies is essential for enhancing the safety, reliability, and economic efficiency of transportation infrastructure, particularly roads and bridges.Conventional optimization models typically aim to minimize costs while maximizing reliability but often overlook the varied risk attitudes of infrastructure managers and decision-makers.This limitation can lead to theoretically optimal strategies that fail to reflect practical stakeholder preferences under uncertainty.This study proposes a framework that integrates risk attitudes directly into the optimization of maintenance strategies.The approach extends a value iteration algorithm by incorporating a Constant Relative Risk Aversion (CRRA) utility function, enabling the modeling of risk-averse, risk-neutral, and riskseeking behaviors.The framework is implemented with an actor-critic reinforcement-learning architecture, where a critic network guides a policy (actor) toward cost-optimal maintenance decisions.Unlike traditional reinforcement learning or optimization methods that assume risk neutrality, the proposed model adjusts the reward structure to reflect how decision-makers perceive the trade-offs between cost, reliability, and future uncertainty.The framework is applied to a representative road and bridge network to examine how varying risk preferences influence the selection of optimal maintenance strategies-including preventive, corrective, opportunistic, and induced maintenance.By adjusting the risk aversion parameter, the model produces different optimal strategies corresponding to diverse decisionmaker profiles.To support decision-making, the study visualizes how risk attitudes affect both the selection of maintenance strategies and the associated lifecycle costs.Results highlight the significant role of risk preferences in shaping maintenance policies.This research contributes to bridging the gap between theoretical optimization techniques and the practical realities of infrastructure management by providing a flexible, risk-aware tool that enables infrastructure managers to tailor maintenance strategies to both technical requirements and organizational risk preferences.
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 | 0.002 | 0.005 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.002 | 0.002 |
| 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".