•Create a database from more than 141 million AIS records.•Propose a random forest-based model to measure the similarity between trajectories.•Develop a decision strategy for vessel destination ...prediction.
Shipping is one of the major transportation approaches around the world. With the growing demands for global shipping service, vessel destination prediction has shown its significant role in improving the efficiency of decision making in industry and ensuring a safe and efficient maritime traffic environment. Currently, most vessel destination prediction methods focus on regional destination prediction, which has restrictions on destinations and regions. Thus, this paper proposes a general AIS (Automatic Identification System) data-driven model for vessel destination prediction. In this random forest-based model, the similarity between the vessel’s traveling and historical trajectories are measured and utilized to predict the destination. The destination of the historical trajectory, which shares the highest similarity with the traveling trajectory, is predicted as the vessel’s destination. The method is different from previous work which used maritime records as input to predict the destination. In our method, a historical trajectory database was generated from more than 141 million AIS records, which covers 534,824 traveling patterns between ports and more than 5.9 million historical trajectories. Comparative studies were carried out to validate the performance of the proposed model, where eleven state-of-the-art trajectories similarity measurement methods combined with two different decision strategies were implemented and compared. The experimental results demonstrate that the proposed model combined with the port frequency-based decision strategy achieves the best prediction accuracy on 35,937 testing trajectories.
In recent years, many vessels have utilized Automatic Identification Systems (AIS) to process marine positioning and navigation; the system employs GPS to position and transmits signals through ...ultra-high frequency wireless communication, which can deliver the signals to a broader range. However, due to the high power-consumption character of AIS, it is necessary to consider an appropriate power supply to support the system. In other words, AIS is suitable for various application fields, but the power supply issue remains unresolved in different Internet of Things (IoT) equipment, such as marine movement or land activity positioning. This article proposes a sensor designed with a self-powered AIS, and the primary functions are: 1. Available for customizing the signal sending time intervals based on the site needs; 2. A panic button is designed in the system for activating AIS signal transmission; 3. The equipment stores electricity in solar panels to improve the power consumption issue of AIS; 4. The g-sensor installed in the equipment detects and judges AIS motion and wobble to position the vessel. The AIS equipment presented in this research mainly applies to positioning marine and mountain activities or ocean farming. Due to the lack of power supply in the applications mentioned above, if AIS cannot deliver signals for a long time, the equipment will influence the power issue of other detectors in the entire design. Therefore, this study discusses a design combined with a self-powered AIS detector, allowing the AIS equipment to operate for a year and be suitable for various applications. This design can resolve the lack of mobile communication issues on the sea or in the mountain.
To achieve reliable mining results for massive vessel trajectories, one of the most important challenges is how to efficiently compute the similarities between different vessel trajectories. The ...computation of vessel trajectory similarity has recently attracted increasing attention in the maritime data mining research community. However, traditional shape- and warping-based methods often suffer from several drawbacks such as high computational cost and sensitivity to unwanted artifacts and non-uniform sampling rates, etc. To eliminate these drawbacks, we propose an unsupervised learning method which automatically extracts low-dimensional features through a convolutional auto-encoder (CAE). In particular, we first generate the informative trajectory images by remapping the raw vessel trajectories into two-dimensional matrices while maintaining the spatio-temporal properties. Based on the massive vessel trajectories collected, the CAE can learn the low-dimensional representations of informative trajectory images in an unsupervised manner. The trajectory similarity is finally equivalent to efficiently computing the similarities between the learned low-dimensional features, which strongly correlate with the raw vessel trajectories. Comprehensive experiments on realistic data sets have demonstrated that the proposed method largely outperforms traditional trajectory similarity computation methods in terms of efficiency and effectiveness. The high-quality trajectory clustering performance could also be guaranteed according to the CAE-based trajectory similarity computation results.
The flowchart of our proposed CAE-based unsupervised learning method for vessel trajectory similarity computation and its application to vessel trajectory clustering. Display omitted
•Similarities between vessel trajectories are equivalent to computing similarities between the informative trajectory images.•An unsupervised learning method is proposed to measure similarities between different informative trajectory images.•The proposed learning method shows superior performance in terms of both efficiency and effectiveness.
•Static risk modelling of individual ships based on Bayesian networks.•Ship Risk Profile parameters of Port State Control inspection used as risk variables.•Characterization of static risk profile of ...the maritime traffic.•Static risk profile of individual ships under incomplete information.
This paper proposes a probabilistic approach for characterising the static risk of individual ships based on Bayesian networks (BNs). The approach uses the Ship Risk Profile parameters of the New Inspection Regime of the Paris Memorandum of Understanding (MoU) on Port State Control (PSC), not as risk factors for ship selection in PSC inspections, but as risk variables for ship risk assessment and maritime traffic monitoring. The objectives of the proposed approach are threefold: the characterisation of the static risk profile of the maritime traffic crossing a given geographic area; the identification of the most likely circumstances under which a specific static risk profile is expected to occur; and the characterisation of the static risk profile of individual ships in the presence of incomplete information, such as that obtained from the Automatic Identification System. A dataset collected from the Paris MoU platform is used for the development of the BN model and its validity is assessed. A quantitative assessment for the predictive validity of the model is also conducted by a sensitivity analysis that shows the consistency of the model with the Ship Risk Profile criteria and with the results of other studies developed also from historical PSC inspection data.
Automatic identification system (AIS) can provide massive ship trajectory data that is valuable for mining information in water traffic. However, large sizes lead to difficulties in storing, ...querying, and processing the aforementioned data. In the present study, to better compress ship trajectory data regarding compression time and efficiency, a method based on the improved Douglas–Peucker (DP) algorithm is presented. In the process of compression, the proposed method considers the shape of vessel trajectory derived from course information of track points. Parallel experiments are conducted based on AIS data gathered over the duration of a month in the Chinese Zhou Shan islands. The results indicate that this method can effectively compress ship trajectory information. Additionally, when compared with the traditional DP algorithm, this method can significantly reduce the compression time and exhibits better performance at high compression strengths. Also, the proposed method outperforms other existing trajectory compression algorithms in term of compression time.
∙This study proposes a method that incorporates the course change in the ship trajectory and DP algorithm.∙Parallel experiments are conducted based on AIS data gathered over the duration of a month in the Chinese Zhou Shan islands. The results indicate that this method can effectively compress ship trajectory information.∙Additionally, when compared with the traditional DP algorithm, this method can significantly reduce the compression time and exhibits better performance at high compression strengths.∙Also, the proposed method outperforms other existing trajectory compression algorithms (OPW; TD-TR; OPW-TR) in term of compression time.
Large marine protected areas (MPAs) have recently been established throughout the world at an unprecedented pace, yet the value of these reserves for mobile species conservation remains unclear. Reef ...shark populations continue to decline even within some of the largest MPAs, fueling unresolved debates over the ability of protected areas to aid mobile species that transit beyond MPA boundaries. We assessed the capacity of a large MPA to conserve grey reef sharks - a Near Threatened species with a widespread distribution and poorly understood offshore movement patterns - using a combination of conventional tags, satellite tags, and an emerging vessel tracking technology. We found that the 54,000km2 U.S. Palmyra Atoll National Wildlife Refuge in the central Pacific Ocean provides substantial protection for grey reef sharks, as two-thirds of satellite-tracked sharks remained within MPA boundaries for the entire study duration. Additionally, our analysis of >0.5 million satellite detections of commercial fishing vessels identified virtually no fishing effort within the refuge and significant effort beyond the MPA perimeter, suggesting that large MPAs can effectively benefit reef sharks and other mobile species if properly enforced. However, our results also highlight limitations of place-based conservation as some of these reef-associated sharks moved surprising distances into pelagic waters (up to 926km from Palmyra Atoll, 810km beyond MPA boundaries). Small-scale fishermen operating beyond MPA boundaries (up to 366km from Palmyra) captured 2% of sharks that were initially tagged at Palmyra, indicating that large MPAs provide substantial, though incomplete, protection for reef sharks.
•We used two tag types and vessel tracking to assess a large marine protected area.•Large MPAs offer substantial, though not absolute, protection for reef sharks.•Most grey reef sharks stayed within the MPA near minimal observed fishing effort.•Two swam past the largest possible MPA limits (up to 926km into open water).•2% of sharks tagged in the MPA were caught on atolls 223–366km away from the MPA.
Predicting the likelihood of maritime accidents is hindered by the relative sparsity of collisions on which to develop risk models. Therefore, significant research has investigated the capability of ...non-accident situations, near misses and encounters between vessels as a surrogate indicator of collision risk. Whilst many studies have developed ship domain concepts, few have considered the practical considerations of implementing this method to characterise navigational risk between waterways and scenarios. In order to address this, within this paper we implement and evaluate the capability and validity of domain analysis to characterise and predict the likelihood of ship collisions. Our results suggest that the strength of the relationship between collisions and encounters is varied both between vessel types and the spatial scale of assessment. In addition, we demonstrate some key practical considerations in utilising domain analysis to predict the change in collision risk, through a hypothetical wind farm. The outcomes of this study provide research direction for practical applications of domain analysis on collision risk assessments.
•The suitability of domain analysis for collision risk assessment is discussed.•A national picture of encounter frequency is used to inform collision risk.•The relationship between encounters and historical collisions is investigated.•A practical method for predicting the change in collision risk is proposed.
With the establishment of satellite constellations and terrestrial networks of Automatic Identification System (AIS) receivers, an increasing number of ship trajectories have become available, and ...the data size of trajectories that must be recorded is increasing. As a result, transmitting, processing and storing data have become important issues. At the same time, ship behaviour information is hidden in AIS data. Hence, an effective method is required to not only compress redundant information but also maintain the main characteristic elements included in the trajectory. In this paper, a novel algorithm considering the spatial and motion features of trajectories is designed, which can compress AIS trajectories based on ship behaviour characteristics. The proposed algorithm has two main parts: the Douglas-Peucker (DP) algorithm is employed to simplify trajectories according to spatial features, and a sliding window is adopted to simplify trajectories based on motion features. Furthermore, statistical theory is applied to help determine the thresholds of motion features in sliding window algorithms. The two results are merged to form a trajectory simplification algorithm that considers ship behaviours. To verify the effectiveness of the proposed algorithm, numerical experiments are performed. The results indicate that the proposed algorithm can efficiently simplify trajectories by considering ship behaviour as needed.
•This study proposes a simplification method that considers ship behaviours.•Statistical theory is applied to help determine the thresholds of the motion features.•Numerical experiments are performed to verify that the proposed algorithm performance outperforms other methods.
The maritime industry plays a key role in reducing greenhouse gas (GHG) emissions, as an effort to combat the global issue of climate change. The International Maritime Organization (IMO) is ...targeting a 50% reduction in GHG emissions by 2050 compared to 2008. To measure Singapore's progress towards this target, we have conducted a comprehensive analysis of carbon dioxide (CO2) emissions from the Western Singapore Straits based on the voyage data from Automatic Identification System (AIS) and static information from Singapore Maritime Data Hub (SG-MDH). Two methodologies, the MEET and TRENDS frameworks were applied to estimate the emission volume per vessel per hour. The data analysis results were next aggregated and visualised to answer key questions such as: How did the carbon emission level change from 2019 to 2020, in general, and for specific vessel types? What are the top vessel types and flags that had the highest carbon emissions? Did the traffic volume and emission level decrease during the Circuit Breaker period in 2020? The results of this study can be used to review Singapore's emission control measures and will be of value to the Maritime and Port Authority (MPA) of Singapore responsible for managing CO2 emissions at the Singapore Port.
Display omitted
•The MEET and TRENDS framework was used to estimate the CO2 emission per vessel.•Singapore flag vessel CO2 emissions reduced by 10.68% from 2019 to 2020.•Emission level dropped significantly in August 2020 and remained low until end of the year.•Carbon emissions for ferries reduced considerably during the Circuit Breaker.•This study contributes to the management of CO2 emissions at the Singapore Port.
•Empirical evidence of 141 incidences of past port disruptions across 27 events.•Ports disruptions have a median of 6 days with a 95th quantile of 22.2 days.•Ten day disruption in U.S.A. associated ...with a 35 m/s wind speed and/or 2.5 m storm surge.•All events cause simultaneous disruption at multiple ports.•Production recapture more likely than port substitution.
Ports are located in low-lying coastal and riverine areas making them prone to the physical impacts of natural disasters. The consequential disruptions can potentially propagate through supply chains, resulting in widespread economic losses. Previous studies to quantify the risks of port disruptions have adopted various modelling assumptions about the resilience of individual ports and marine network logistics. However, limited empirical evidence is available to validate these modelling assumptions or to provide deeper understanding of the ways in which operations are adapted during and after disruptions. Here, we use vessel tracking data to analyse past port disruptions due to natural disasters, evaluating 141 incidences of disruptions across 74 ports and 27 disasters. Results show a median disruption duration of six days with a 95th percentile of 22.2 days. All analysed events show multiple ports being affected simultaneously, challenging some of the studies that only focus on single port disruptions. Moreover, we find that the duration of the disruption scales with the severity of the event, with an increment of 1.0 m storm surge or 10 m/s wind speed associated with a two day increase in disruption duration. In contrast to commonplace assumptions in model studies, substitution between ports is rarely observed during short-term disruptions. On the other hand, production recapture happens in practice in many cases of port disruptions. In short, empirical vessel tracking data provides valuable insights for future modelling studies in order to better approximate the extent of the disruption and the potential resilience of the port and maritime network.