A Quantum Support Vector Regression Framework for Missing Data Imputation in Ship Performance Data
Organizations: HPC & Quantum, Bull Spain, Madrid, 28040, Spain · ARIES Research Group, Nebrija University, Madrid, 28015, Spain · Innovation & Technology Department, Laskaridis Shipping Co. Ltd, Athens, 14562, Greece
Abstract
The Internet of Ships enables vessels to generate vast amounts of data, yet missing values remain a persistent issue. To the best of the authors' knowledge, existing imputation techniques in this domain rely exclusively on classical machine learning. This study investigates quantum support vector regression (QSVR) for imputing missing values in ship performance data. Classical support vector regression and QSVR models based on different quantum feature maps are compared for estimating marine fuel consumption under varying levels of data availability. The case study uses real-world data from a Panamax bulk carrier, covering 37 variables across six categories. Three experiments are conducted: exploratory analysis with three input features, scalability analysis with increasing dimensionality (3 to 18 features), and comparative evaluation under a selected representative configuration. Across all scenarios, both classical and quantum approaches captured the relationship between the inputs and the main-engine fuel-flow rate, maintaining strong imputation performance even under severe data scarcity. ZZFeatureMap_1 achieved the highest average R2 (0.991 in the performance comparison conducted with 60% of the dataset, 90% missing data, and 14 input features, for instance) and the lowest aggregate imputation errors, while the radial basis function kernel was the strongest classical baseline, delivering highly competitive results.
Figures & tables
| Reference | Data Imputation Method | Type of Data |
|---|---|---|
| Lu et al. (2026) | GAN-based method | Noon reports and AIS data |
| Xuan et al. (2026) | Graph-enhanced Transformer | AIS data |
| Çelik (2026) | Fourier-based Compressed Sensing (F-CS) model | Buoy data |
| Wang et al. (2025) | Conditional diffusion | AIS data |
| Zhu et al. (2025) | Random Forest | Ship maintenance data |
| Muhammed Nor et al. (2025) | MICEForest, kNN, and MissForest | Environmental Performance Index (EPI) data |
| Parameter | Value | Unit |
|---|---|---|
| Vessel type | Panamax Bulk Carrier | – |
| Deadweight (DWT) | 75,200 | t |
| Length overall (LOA) | 225.00 | m |
| Length between perpendiculars (LBP) | 217.00 | m |
| Rule length | 213.885 | m |
| Breadth (moulded) | 32.26 | m |
| Category | Variable | Description | Unit | Source |
| Main Engine | Main Engine Rotational Speed | Main engine crankshaft rotational speed | rpm | Control Alarm Monitoring System |
| Main Engine Fuel Load | Main engine load | % | Fuel Load Transducer | |
| Main Engine Scavenging Air Pressure | Main engine scavenging air pressure | bar | Pressure Transducer | |
| Main Engine Turbocharger Rotational Speed | Main engine turbocharger rotational speed | rpm | RPM Transducer | |
| Main Engine Turbocharger Rotational Speed | Main engine turbocharger rotational speed | rpm | RPM Indicator | |
| Main Engine Fuel Oil Inlet Mass Flow | Main engine fuel oil inlet mass flow | kg h -1 | Torquemeter |
| Missing values | Train size | Test size |
|---|---|---|
| 10% | 12,644 | 1,405 |
| 20% | 11,239 | 2,810 |
| 30% | 9,834 | 4,215 |
| 40% | 8,429 | 5,620 |
| 50% | 7,025 | 7,024 |
| 60% | 5,620 | 8,429 |
| Rank | Variable | Reason for relevance to fuel flow |
|---|---|---|
| 1 | Main Engine Fuel Load (%) | Most direct representation of engine operating demand and therefore expected fuel requirement |
| 2 | Propeller Shaft Mechanical Power (kW) | Direct measure of delivered propulsion power, strongly physically related to fuel consumption |
| 3 | Propeller Shaft Torque (N m) | Directly associated with engine/shaft loading and delivered power |
| 4 | Main Engine Rotational Speed (rpm) | Fundamental engine operating parameter governing power and fuel demand |
| 5 | Main Engine Scavenging Air Pressure (bar) | Strong indicator of engine load and combustion operating condition |
| 6 | Main Engine Turbocharger Rotational Speed, RPM Transducer | Closely responds to engine load, exhaust energy, and combustion demand |
Incomplete Data Imputation
Incomplete data imputation estimates unobserved values while accounting for data structure, missingness patterns, and uncertainty.
See all 82 papers