MASS Remote Operation Centres — State of the Art (CMOROC Appendix B)¶

Status: established
Last updated: 2026-05-31
Sources: Cmoroc_Appendix B_State Of The Art_V2 2_231025.Pdf
Tags: [cmoroc, emsa, mass, roc, state-of-the-art, shore-control-centre, munin, autoship, safemass, remote-tower, cross-domain, situational-awareness, cybersecurity]

Summary¶

Appendix B of the EMSA CMOROC study (European Maritime Safety Agency, 2023b) reviews the state of the art in Remote Operation Centres (ROCs) for autonomous shipping. It surveys EMSA's own prior studies (SAFEMASS, RBAT), the Maritime UK Code of Practice, eight maritime projects (MUNIN, H2H, AUTOSHIP, AVATAR, LOAS, Seafar, FernBin, DFFAS) and two industry programmes (Rolls-Royce, Kongsberg), then looks across domains to aviation remote-tower and RPAS control and to rail. Each project is examined under four recurring questions — automation use cases, ROC organisation, safety and security, and legal aspects — and each yields tagged "statements and findings" in four categories (C1 principal MASS/ROC elements, C2 technical requirements, C3 human-interaction requirements, C4 legal requirements) that feed the rest of the CMOROC study.

Body¶

Context¶

Appendix B (European Maritime Safety Agency, 2023b) is the evidence-gathering stage of the CMOROC study: a curated, not exhaustive, review of how ROCs have been conceived and built across maritime and adjacent domains, conducted by the study team (Hochschule Bremen, Humatects, DLR) for EMSA. It supplies the empirical grounding for the process map, roles, and competence tables synthesised in Cmoroc Roc Competence Framework, and its findings are the input the later appendices convert into requirements. Within this knowledge base it is the systematic survey behind the abstract ROC concept of Remote Operation Centres Mass — it names and compares the very projects (MUNIN, AUTOSHIP) that article treats — and it independently reviews the QGILD automation-transparency work analysed in Human In The Loop Automation Transparency. It is one of the three formerly-deferred CMOROC PDFs now compiled.

Key Points¶

The review frames a single design problem. A ROC is an onshore centre for monitoring and controlling one or more ships, the operators physically separated from the vessels; the report treats Shore Control Centre, Shore Operations Centre, Remote Control Centre and Base Control Station as synonyms (PDF p. 8, orig. p. 6). Two project archetypes recur: advanced concepts (MUNIN, LOAS) where monitoring is normal and human control is only needed when onboard automation approaches its limits, and direct-steering concepts (Seafar, FernBin) where the ROC remotely steers a vessel with limited onboard automation (PDF p. 8, orig. p. 6). The governing question is how to keep the operator informed and able to intervene, or to bring them back into the loop quickly when needed (PDF p. 8, orig. p. 6). Because guidelines for designing and certifying a ROC are absent, early projects simply replicated the ship bridge ashore, raising the open question of how much "ship sense" (smell, sound, vibration, motion) must be reproduced versus replaced by more reliable sensors (PDF p. 8, orig. p. 6, citing Dybvik, Veitch & Steinert, 2020; Porathe, 2021). The report poses the design questions the rest of CMOROC answers: which ROC roles are needed and how they cooperate, which competencies they require, how many ships one operator can monitor, how to support situational awareness through the human–machine interface, and when operators should take manual control (PDF p. 8, orig. p. 6).

EMSA's own prior studies set the regulatory baseline. SAFEMASS (DNV GL, July 2019 – March 2020) identified emerging risks and regulatory gaps from MASS introduction, working from IMO's degrees of autonomy (PDF p. 9, orig. p. 7). RBAT, ongoing at the time of writing, developed a five-part Risk-Based Assessment Tool — describe the use of automation and remote control, perform hazard analysis, perform mitigation analysis, perform risk evaluation, and address risk control (PDF p. 9, orig. p. 7).

The maritime project review produces structured findings. MUNIN is the most developed precedent: it proposes a Shore Control Centre with four roles — ROC Operator, SCC Supervisor, SCC Captain, SCC Engineer (a C1 finding) — a hierarchical status display with coloured flags, spatial and temporal voyage overviews, time-to-maintenance versus time-to-destination displays, and trendlines for anomaly detection (C3 findings), plus eleven risk control options (C2) and explicit task allocations for safe speed, radar/ECDIS/AIS/GNSS/ARPA use, and lookout (C4) (PDF pp. 24–25, orig. pp. 22–23). AUTOSHIP introduces the Operational Design Domain (ODD), borrowed from automotive, to formalise the conditions under which the automation is designed to operate, and structures the architecture around a Minimum Risk Condition unit that computes a minimum-risk manoeuvre alongside the ROC Operator (PDF pp. 28–33, orig. pp. 27–32, citing Rødseth, Lien Wennersberg & Nordahl, 2022; Bolbot et al., 2021). Across projects the report records risk control options such as ensuring sufficient ROC supervision capacity and ensuring the MASS fleet can enter minimum-risk conditions (PDF p. 33, orig. p. 31).

Human-element risk is treated systematically. The review documents three human-in-the-loop risk categories drawn from the project HAZID and fault-tree analyses: inability to build situational awareness, mode confusion, and (dis)trust in automation (PDF pp. 28–30, orig. pp. 27–29). The QGILD work (Porathe, 2021, 2022) is reviewed specifically for automation transparency and quickly getting the operator back into the loop (PDF pp. 38–39, orig. pp. 37–38) — the same display analysed in Human In The Loop Automation Transparency. The recurring conclusion across maritime projects is that one operator overseeing several ships in normal operation is the expected future, and that the ROC operator's role resembles but exceeds a Vessel Traffic Service operator's, because the ROC operator may take direct control rather than only monitor and assist (PDF p. 8, orig. p. 6).

The cross-domain review imports lessons from aviation and rail. Because remote control is new to shipping, the report examines aviation remote-tower and RPAS projects (INVIRCAT, DTT, CORUS) and a rail project (ARTE) under the same four questions (PDF pp. 56–77, orig. pp. 54–75). INVIRCAT addresses IFR RPAS control in terminal airspace with no crew aboard and an Automatic Take-Off and Landing system, producing a concept of operations and recommendations for rule-makers (PDF p. 56, orig. p. 54). The remote-tower and U-space material supplies models for one-to-many supervision, control-room organisation, and the staged levels of automation that the maritime ROC concept lacks a settled equivalent for (PDF pp. 60–67, orig. pp. 58–65, citing Friedrich, Timmermann & Jakobi, 2022; SESAR Joint Undertaking, 2017).

Conclusion¶

Appendix B concludes that ROC design lacks accepted guidelines and that the field is converging on a small set of shared problems rather than a shared solution: defining ROC roles and their cooperation, supporting situational awareness and re-entry into the loop, allocating control between automation and operator, bounding the automation with an ODD and minimum-risk conditions, and securing the system against cyber threats. The review's value to CMOROC is its structured output — every project distilled into C1–C4 findings — which the study then converts into the process map, roles, and competence tables of Cmoroc Roc Competence Framework. The cross-domain comparison underlines that maritime remote operation is less mature than aviation remote-tower control, and that the one-operator-many-ships model treated in Remote Operation Centres Mass is the assumed direction of travel but is not yet regulated.

References¶

Bolbot, V., Theotokatos, G., Boulougouris, E. & Vassalos, D. (2021) 'A novel cyber-risk assessment method for ship systems', Safety Science, 131, 104908. doi: 10.1016/j.ssci.2020.104908. To be validated.

Dybvik, H., Veitch, E. & Steinert, M. (2020) 'Exploring challenges with designing and developing shore control centers (SCC) for autonomous ships', Proceedings of the Design Society: DESIGN Conference, 1, pp. 847–856. doi: 10.1017/dsd.2020.131. To be validated.

European Maritime Safety Agency (2023b) CMOROC Appendix B - State of the Art. Identification of Competences for MASS Operators in Remote Operation Centres, V2.2. Lisbon: EMSA. cmoroc2023appendixB

Friedrich, M., Timmermann, F. & Jakobi, J. (2022) 'Concept of operation for remote tower control of multiple airports', in Proceedings of the Human Factors and Ergonomics Society Europe Chapter Annual Meeting. To be validated.

Porathe, T. (2021) 'Maritime autonomous surface ships and the COLREGs: Do we need quantitative or qualitative rules?', Proceedings of the AHFE Conference. To be validated.

Porathe, T. (2022) 'Remote monitoring of autonomous ships: A quickly getting into the loop display (QGILD)', Advances in Transportation, 60, pp. 691–697. doi: 10.54941/ahfe1002506. porathe2022qgild

Rødseth, Ø.J., Lien Wennersberg, L.A. & Nordahl, H. (2022) 'Towards approval of autonomous ship systems by their operational envelope', Journal of Marine Science and Technology, 27, pp. 67–76. doi: 10.1007/s00773-021-00815-z. To be validated.

SESAR Joint Undertaking (2017) U-space blueprint. Brussels: SESAR Joint Undertaking. To be validated.

Open Questions¶

How much "ship sense" must a ROC reproduce, and where do new sensors outperform replicated human senses? (Dybvik, Veitch & Steinert, 2020, as reviewed.)
Which of the aviation remote-tower staffing and one-to-many models transfer to maritime ROCs given the ROC operator's direct-control role?