Application of reinforcement learning for integrating project risk analysis and risk response planning: A case study on construction projects
Bibliographic record
Abstract
Project Risk Management contains processes ranging from planning to control. It is applied to identify risks, analyze them, and design responses to change the occurrence rate and/or the effect of project risks. It is important for project managers to analyze the effects of the risks in projects and also consider project risks in their decisions. If project risks are not addressed during the risk management process, issues such as schedule delays, cost overruns, and even project failure may occur. This paper aims to introduce a Markov method to integrate project risk analysis and risk response planning. This method is applied to forecast the following status of the project when limited information about the project is available. Moreover, earned value management (EVM) methods were used to include various types of project risks through the project lifecycle. The model also offers the capability to choose the most effective risk response for managing project risks through the application of the Markov decision process (MDP). Eventually, we introduce a case study to demonstrate functionality and effectiveness of the presented approach. Solving the model allows for identifying the best set of risk response strategies tailored to each specific project state. The computational results illustrate that the current state of the project has a significant impact in the process of risk response planning. Since uncertainty is the inherent characteristic of projects, the use of the project’s current state is more reliable than the previous status of projects, and the Markov method is applied in this research because it uses the current state for its modelling. Using this method, managers can predict the future state of projects and find the best response in each status of projects.
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 | 0.003 | 0.005 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.002 | 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".