CoChat

Analyzing Operator States and the Impact of AI-Enhanced Decision Support in Control Rooms: A Human-in-the-Loop Specialized Reinforcement Learning Framework for Intervention Strategies

Open access

The DOI matches its Crossref record; 42 of the 42 references it cites were found in a registry or an index.

Abstract

In complex industrial and chemical process control rooms, effective decision-making is crucial for safety and efficiency. The experiments in this paper evaluate the impact and applications of an AI-based decision support system integrated into an improved human-machine interface, using dynamic influence diagrams, a hidden Markov model, and deep reinforcement learning. The enhanced support system aims to reduce operator workload, improve situational awareness, and provide different intervention strategies to the operator adapted to the current state of both the system and human performance. Such a system can be particularly useful in cases of information overload when many alarms and inputs are presented all within the same time window, or for junior operators during training. A comprehensive cross-data analysis was conducted, involving 47 participants and a diverse range of data sources such as smartwatch metrics, eye-tracking data, process logs, and responses from questionnaires. The results indicate interesting insights regarding the effectiveness of the approach in aiding decision-making, decreasing perceived workload, and increasing situational awareness for the scenarios considered. Additionally, the results provide insights to compare differences between styles of information gathering when using the system by individual participants. These findings are particularly relevant when predicting the overall performance of the individual participant and their capacity to successfully handle a plant upset and the alarms connected to it using process and human-machine interaction logs in real-time which resulted in a 95.8% prediction accuracy using hidden Markov model. These predictions enable the development of more effective intervention strategies.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1080/10447318.2024.2391605

Cites

42 of 42 references verified0 citation errors · 0 not checkable

Statistics

Figures

Image 1
CC-BY · From the PDF
Image 2
CC-BY · From the PDF
Image 3
CC-BY · From the PDF
Figure 1 Case study: simulated control room environment of a process indus­ try for a living lab.
Figure 1 Case study: simulated control room environment of a process indus­ try for a living lab.CC-BY · From the PDF
Figure 2 Specialized reinforcement learning agent (SRLA). Source : Abbas et al. (2024).
Figure 2 Specialized reinforcement learning agent (SRLA). Source : Abbas et al. (2024).CC-BY · From the PDF
Figure 3 Human-centered specialized reinforcement learning agent for safety-critical systems. The dashed area represents the framework developed in Mietkiewicz et al. (2023, 2024).
Figure 3 Human-centered specialized reinforcement learning agent for safety-critical systems. The dashed area represents the framework developed in Mietkiewicz et al. (2023, 2024).CC-BY · From the PDF
Image 7
CC-BY · From the PDF
Image 8
CC-BY · From the PDF
Image 9
CC-BY · From the PDF
Figure 7 Specialized reinforcement learning agent training and inference for process abnormality related to (a) tank pressure and (b) reactor temperature.
Figure 7 Specialized reinforcement learning agent training and inference for process abnormality related to (a) tank pressure and (b) reactor temperature.CC-BY · From the PDF
Figure 8 Dataset file hierarchy.
Figure 8 Dataset file hierarchy.CC-BY · From the PDF
Image 12
CC-BY · From the PDF
Image 13
CC-BY · From the PDF
Figure 10 Questions related to deep reinforcement learning suggestions.
Figure 10 Questions related to deep reinforcement learning suggestions.CC-BY · From the PDF
Figure 11 Heat maps for participants in (a) GroupN (without AI support) and (b) GroupAI (with AI support).
Figure 11 Heat maps for participants in (a) GroupN (without AI support) and (b) GroupAI (with AI support).CC-BY · From the PDF
Image 16
CC-BY · From the PDF
Figure 12 Heat maps for participants in (a) GroupN (without AI support) and (b) GroupAI (with AI support) within the time of Interest (TOI) of critical alarms.
Figure 12 Heat maps for participants in (a) GroupN (without AI support) and (b) GroupAI (with AI support) within the time of Interest (TOI) of critical alarms.CC-BY · From the PDF
Figure 13 Radar plot between GroupN (blue) and GroupAI (red).
Figure 13 Radar plot between GroupN (blue) and GroupAI (red).CC-BY · From the PDF
Figure 14 Correlation matrix for participants in both groups.
Figure 14 Correlation matrix for participants in both groups.CC-BY · From the PDF
Figure 15 Significant correlation matrix with values greater than 0.4 and based on two-sample independent t -test and Bonferroni correction.
Figure 15 Significant correlation matrix with values greater than 0.4 and based on two-sample independent t -test and Bonferroni correction.CC-BY · From the PDF
Figure 16 Scree plot and the best possible value of the factors.
Figure 16 Scree plot and the best possible value of the factors.CC-BY · From the PDF
Figure 17 Cumulative explained variance per each factor.
Figure 17 Cumulative explained variance per each factor.CC-BY · From the PDF
Image 23
CC-BY · From the PDF
Image 24
CC-BY · From the PDF
Image 25
CC-BY · From the PDF
Image 26
CC-BY · From the PDF
Image 27
CC-BY · From the PDF
Figure 20 Radar plot between participant 21 and 24..
Figure 20 Radar plot between participant 21 and 24..CC-BY · From the PDF
Image 29
CC-BY · From the PDF
Figure 22 Radar plot between participant 30 and 32.
Figure 22 Radar plot between participant 30 and 32.CC-BY · From the PDF
Image 31
CC-BY · From the PDF
Figure 24 Radar plot between participant 37 and 96.
Figure 24 Radar plot between participant 37 and 96.CC-BY · From the PDF
Figure 25 Frequency plot of consequences of participant's actions during plant abnormality.
Figure 25 Frequency plot of consequences of participant's actions during plant abnormality.CC-BY · From the PDF
Figure 26 Radar plot between GroupN (blue) and GroupAI (red) for scenarios (a) critical alarm and (b) alarm overflow.
Figure 26 Radar plot between GroupN (blue) and GroupAI (red) for scenarios (a) critical alarm and (b) alarm overflow.CC-BY · From the PDF
Image 35
CC-BY · From the PDF
Figure 27 Radar plot between scenarios critical alarm (green) and alarm overflow (violet) for (a) GroupN and (b) GroupAI.
Figure 27 Radar plot between scenarios critical alarm (green) and alarm overflow (violet) for (a) GroupN and (b) GroupAI.CC-BY · From the PDF
Figure 28 Radar plot between scenarios critical alarm (green) and alarm overflow (violet) for participants (a) P21, (b) P24, (c) P32, and (d) P37.
Figure 28 Radar plot between scenarios critical alarm (green) and alarm overflow (violet) for participants (a) P21, (b) P24, (c) P32, and (d) P37.CC-BY · From the PDF
Image 38
CC-BY · From the PDF
Figure 30 Top 10 factor loadings in both directions on the Principal Component analysis (PCA).
Figure 30 Top 10 factor loadings in both directions on the Principal Component analysis (PCA).CC-BY · From the PDF
Image 40
CC-BY · From the PDF
Image 41
CC-BY · From the PDF
Image 42
CC-BY · From the PDF

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 42 references

Research works printed but not linked: 3

  1. 10.1016/j.dib.2024.110170 · DOI
  2. Lee, A. (2023). hmmlearn: Hidden Markov models in python with Scikit-learn like API. Python Package Index.
  3. 10.48550/arXiv.2004.05490 · arXiv · doi:10.48550/arxiv.2004.05490

Cited by

See all 37

Metadata

Authors
Abbas, A. N., Amazu, C. W., Mietkiewicz, J., Briwa, H., Perez, A. A., Baldissone, G., Demichela, M., Chasparis, G. C., Kelleher, J. D., & Leva, M. C.
Affiliations
  • Technological University Dublin (IE)
  • Politecnico di Torino (IT)
  • Software Competence Center Hagenberg (Austria) (AT)
  • Trinity College Dublin (IE)
Corresponding author
Abbas, Ammar N.
Venue
International Journal of Human–Computer Interaction
Published
2024-09-04
Volume
41
Issue
12
Pages
7218-7252
ISSN
  • 1044-7318 (ISSN-L)
  • 1532-7590
OpenAlex
W4402234944
Cited by
34
References
41
Citations by year
2026: 23 · 2025: 10 · 2024: 1
FWCI
16.13
Type
journal article
Language
en
Open access
Locations
  • International Journal of Human-Computer Interaction (publishedVersion, cc-by)
  • arXiv (Cornell University) (submittedVersion)
  • PORTO@Iris (submittedVersion, cc-by)
  • Arrow - TU Dublin (Technological University Dublin) (submittedVersion, cc-by)
  • arXiv (Cornell University)
Funders
  • Canadian Institute of Steel Construction (955901)
  • European Commission (955901)
Topics
  • Social Sciences › Human Factors and Ergonomics › Human-Automation Interaction and Safety
  • Physical Sciences › Engineering › Safety, Risk, Reliability and Quality › Risk and Safety Analysis
  • Physical Sciences › Computer Science › Human-Computer Interaction › Usability and User Interface Design
Keywords
AI decision support systems, human-machine interaction, control room, human-in-the-loop reinforcement learning, hidden Markov models, operator workload, situational awareness, operator performance prediction, eye tracking, industrial process control, intervention strategies

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Analyzing Operator States and the Impact of AI-Enhanced Decision Support in Control Rooms: A Human-in-the-Loop Specialized Reinforcement Learning Framework for Intervention Strategies. https://app.cochat.ai/papers/10.1080/10447318.2024.2391605
The paper
Abbas, A. N., Amazu, C. W., Mietkiewicz, J., Briwa, H., Perez, A. A., Baldissone, G., Demichela, M., Chasparis, G. C., Kelleher, J. D., & Leva, M. C. (2025). Analyzing Operator States and the Impact of AI-Enhanced Decision Support in Control Rooms: A Human-in-the-Loop Specialized Reinforcement Learning Framework for Intervention Strategies. International Journal of Human–Computer Interaction. https://doi.org/10.1080/10447318.2024.2391605

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.