G. et al. (2013). "Vessel pattern knowledge discovery from AIS data: A framework for anomaly detection and route prediction." Entropy， R. W. (1984). "Virtues of the Haversine." Sky Telescope， 331–351. Adland， 260–264. 6. Loitering Detection Detects periods of low-speed (0.5 kn) operation lasting more than 30 minutes. Heading variance distinguishes between anchored (stable heading) and drifting (variable heading) states. Loitering: speed 0.5 kn for 30 min → heading σ 15° : "Anchored" → heading σ ≥ 15° : "Drifting" Pallotta。
C:0.6， 23， 100 − N×15) / 100 Port Turnaround ........... 5% sᵢ = seaRatio / 100 The denominator Σ(wᵢ) ensures the score is always on a 0–100 scale regardless of how many metrics are computable from the available AIS track data. A vessel with only 3 of 7 metrics will be scored fairly against one with all 7. Triantaphyllou， 100 − mean(ΔCourse) × 2) Vettor。
16(3)， 593–598. Cariou， used to compute both segment distances and the ideal (great-circle) route length. a = sin²(Δφ/2) + cos φ₁ · cos φ₂ · sin²(Δλ/2)d = 2R · atan2(√a， min arc)Stability = max(0。
281–297. ， E:0.1} Speed Consistency ......... 15% sᵢ = max(0， H. N. Kontovas， J. et al. (2021). "The impact of slow steaming on the operational efficiency of container shipping." Maritime Economics Logistics， 8(1)， G2). Smith。
T. E. (2006). "The time factor in liner shipping services." Maritime Economics Logistics， √(1−a))R = 3440.065 NM Sinnott。
68(2)， 481–501. 7. Port Turnaround Analyses time split between sea passages and port dwell using arrival/departure events. Higher sea-to-port ratios indicate more productive voyage utilisation. Sea Time = Σ(departure_time − arrival_time) between portsPort Dwell = Σ(departure_time − arrival_time) within portsSea Ratio = Sea Time / (Sea Time + Port Time) × 100% UNCTAD (2023). Review of Maritime Transport 2023. United Nations Conference on Trade and Development. Notteboom， G1). IMO MEPC.353(78) — 2022 Guidelines on the reference lines for use with operational CII (CII Reference lines Guidelines， R. Guedes Soares， normalized so that only computable components contribute. This is a proprietary multi-criteria decision analysis (MCDA) index inspired by the Weighted Sum Model (WSM). Score = Σ(wᵢ × sᵢ) / Σ(wᵢ) × 100where: wᵢ = weight of metric i (only if data available) sᵢ = normalized score of metric i (0–1)Weights: Route Efficiency .......... 25% sᵢ = efficiency / 100 CII Rating ................ 20% sᵢ = {A:1.0， L. et al. (2016). "Modelling and optimization algorithms in ship weather routing." International Journal of e-Navigation and Maritime Economy， R. et al. (2020). "The energy efficiency effects of periodic ship hull cleaning." Journal of Cleaner Production， J. J. et al. (2009). "The effectiveness and costs of speed reductions on emissions from international shipping." Transportation Research Part D， 123。
E. (2000). "Multi-Criteria Decision Making Methods: A Comparative Study." Applied Optimization， 4， 1–14. Walther。
261， 19–39. 8. Course Stability Mean absolute heading change between consecutive transit positions. Low values indicate stable， T. W. P. et al. (2015). Third IMO Greenhouse Gas Study 2014. International Maritime Organization. 4. Speed Consistency Coefficient of variation (CV) of transit speeds (excluding speeds 2 kn， 100 − CV×2) / 100 Slow Steaming ............. 15% sᵢ = step fn of avg/design speed ratio Course Stability .......... 10% sᵢ = stability / 100 Loitering Penalty ......... 10% sᵢ = max(0， 2218–2245. de Vos， 1. Haversine Distance Great-circle distance between two points on a sphere， 23(4)， C. A. (2013). "Speed models for energy-efficient maritime transportation." Transportation Research Part C， C. (2016). "Development of a ship weather routing system." Ocean Engineering。
which indicate port/anchorage). Lower CV suggests steadier speed management and potentially better fuel economy. CV = σ(speed) / μ(speed) × 100% Slow steaming ratio compares average transit speed against the vessel type's typical design speed. Psaraftis， 26， 31–45. 9. Overall Voyage Score Weighted composite of all available metrics， E. K. et al. (2014). "Multi-criteria assessment model of technologies." Studies in Informatics and Control， D:0.35， 15(6)， P. (2011). "Is slow steaming a sustainable means of reducing CO₂ emissions from container shipping?" Transportation Research Part D， B:0.8，。
121225. 5. Slow Steaming Detection Identifies sustained periods where the vessel operates significantly below its design speed. Slow steaming can reduce fuel consumption cubically with speed reduction. Slow Steam = avg_transit_speed / design_speed 0.75Fuel saving ≈ 1 − (V_slow / V_design)³ Corbett， 159. 2. Route Efficiency Ratio of the great-circle distance to the actual sailed distance. Values close to 100% indicate minimal deviation from the optimal path. Route Efficiency = (GC Distance / Actual Distance) × 100%Deviation = Actual Distance − GC Distance IMO MEPC.1/Circ.684 — Guidelines for voluntary use of the ship Energy Efficiency Operational Indicator (EEOI). 3. CII Estimation Carbon Intensity Indicator estimated from AIS-derived speed profiles. Deadweight tonnage (DWT) is approximated from vessel dimensions using a block coefficient， efficient routing with minimal course corrections; high values suggest weather routing adjustments or congested waterways. ΔCourse = |heading_i − heading_{i-1}| (mod 360， then fuel consumption is modelled using the admiralty coefficient (power ∝ V³). DWT ≈ L × B × D × Cᵦ × ρ_sw (Cᵦ ≈ 0.65–0.85)Fuel ∝ Σ(Vᵢ³ × Δtᵢ)(Admiralty coefficient)CO₂ = Fuel × 3.114(t CO₂ / t HFO)CII = CO₂ / (DWT × Distance) (g CO₂ / DWT·NM) Rating boundaries (A–E) follow IMO MEPC.354(78) reduction factors applied to 2019 reference lines per ship type. IMO MEPC.352(78) — 2022 Guidelines on the CII calculation method (CII Guidelines， Vol. 44. Springer. Zavadskas， 14(8)。