Search

The information-rich vessel movement data provided by the Automatic Identification System (AIS) has gained much popularity over the past decade, during which the employment of satellite-based receivers has enabled wide coverage and improved data quality. The application of AIS data has developed from simply navigation-oriented research to now include trade flow estimation, emission accounting, and vessel performance monitoring. The AIS now provides high frequency, real-time positioning and sailing patterns for almost the whole world's commercial fleet, and therefore, in combination with supplementary databases and analyses, AIS data has arguably kickstarted the era of digitisation in the shipping industry. In this study, we conduct a comprehensive review of the literature regarding AIS applications by dividing it into three development stages, namely, basic application, extended application, and advanced application. Each stage contains two to three application fields, and in total we identified seven application fields, including (1) AIS data mining, (2) navigation safety, (3) ship behaviour analysis, (4) environmental evaluation, (5) trade analysis, (6) ship and port performance, and (7) Arctic shipping. We found that the original application of AIS data to navigation safety has, with the improvement of data accessibility, evolved into diverse applications in various directions. Moreover, we summarised the major methodologies in the literature into four categories, these being (1) data processing and mining, (2) index measurement, (3) causality analysis, and (4) operational research. Undoubtedly, the applications of AIS data will be further expanded in the foreseeable future. This will not only provide a more comprehensive understanding of voyage performance and allow researchers to examine shipping market dynamics from the micro level, but also the abundance of AIS data may also open up the rather opaque aspect of how shipping companies release information to external authorities, including the International Maritime Organization, port states, scientists and researchers. It is expected that more multi-disciplinary AIS studies will emerge in the coming years. We believe that this study will shed further light on the future development of AIS studies. [ABSTRACT FROM AUTHOR]

• Develop a novel data-driven framework for port-skipping identification. • Empirically explore decision preferences of shipping lines for port-skipping. • Reveal various strategic decisions of different shipping lines to heal service disruptions. • The proposed methodology can be cross-applied to identify other disruption recovery behaviors. Service disruption hurts the reliability of liner shipping schedules and the lack of high-quality data impedes research on schedule recovery in liner shipping. With the support of the Automatic Identification System (AIS), a satellite-based tracking system acquiring real-time records of vessels' navigation trajectories worldwide, this study develops a novel data-driven framework to uncover the vessel schedule disruption recovery behavior (i.e., port-skipping). The framework consists of a series of independent yet closely related algorithms, which are in turn tasked with screening port calls, estimating closed-loop routes, measuring similarity between different routes, and identifying port-skipping behavior, respectively. Compared with partial proprietary data from the shipping industry and port authorities, the identification results can provide transparent datasets with wide applications and much easier access publicly. The developed data-driven framework is implemented with vessel trajectory information of around 2,000 container vessels worldwide from January 2016 to December 2020, and the results prove its validity and practical value for liner shipping scheduling management. [ABSTRACT FROM AUTHOR]

• Propose a standardized method to estimate port turnover rate based on satellite data of ocean ships. • Construct berth polygon areas for terminals based on Greatmaps (GMap) visual technique. • Design two tailor-made algorithms to estimate the berthing time of ship in a berthing event. • Assess port turnover rate with aggregate berthing time at a port and its historical port throughput. • Use calculated turnover rate to estimate the monthly throughput of global ports. Accurate assessment of port turnover rate is essential for port operators and shipping carriers to benchmark and improve their operations. This study proposes a standardized method to estimate the port turnover rate based on satellite data of ocean ships. This method can be generalized to accommodate ports of different geographic and operational characteristics with minimum input and running times. To achieve the research objective, we first construct berth polygon areas for terminals based on Greatmaps (GMap) visual technique. Then, two tailor-made algorithms are proposed to estimate the berthing time of ship in a berthing event. Finally, we assess the port turnover rate with aggregate berthing time at a port and its historical port throughput. Assuming that the turnover rate is unchanged in the short term, we can use the estimated turnover to estimate the monthly throughput of global ports. The findings suggest the average Mean Absolute Percentage Error (MAPE) of our estimation is 3.84%. Standardized and high-frequency port statistics are highly valued by the industry but very costly to access. The proposed method makes high-frequency port turnover rate and throughput available for a wide range of users. The statistics and findings will enhance standardization and transparency of port statistics and promote the sustainable development of port industry. [ABSTRACT FROM AUTHOR]

This study proposes high-frequency container port congestion measures based on Automatic Identification System (AIS) data. Vessel movement information of 3,957 container ships from March 2017 to April 2017 is included. The world top 20 container ports' berth and anchorage areas are identified through Density Based Spatial Clustering of Applications with Noise (DBSCAN) and convex hull methods, and their hourly port congestion statuses are depicted in terms of the traffic volume and turnaround time. The constructed congestion measures overcome the disadvantages of the traditionally used port or industry data, which is heterogenous, behind the time and not easy to obtain publicly. The higher frequency (hourly) of the proposed measures can effectively monitor any slight change in port performance. A Long Short-Term Memory (LSTM) neural network model is then proposed for congestion prediction using constructed congestion measures. Point prediction and sequence prediction are both performed. We innovatively introduce congestion propagation effects into the prediction model as input features. Using Shanghai, Singapore and Ningbo ports as case studies, results show that the inclusion of congestion propagation effect can improve the prediction performance especially for sequence prediction. This study provides significant implications and decision support for container shipping market participants. [ABSTRACT FROM AUTHOR]

• We explore the reasons behind the wrong destination reports in AIS. • We use data mining techniques to extract information from AIS. • We build a discrete choice model to analyse the reliability of observed reports. • Our model has an 84.1% accuracy rate in ascertaining the correctness of destination reports. • We provide managerial and policy suggestions to ship operators, port authorities, and regulators. The Automatic Identification System (AIS) is a shipping database that includes the physical characteristics and real-time dynamics of ships. It has attracted great attention from academia recently and has been broadly applied in solving various problems in different fields. The voyage destination report is a piece of information recorded in AIS that indicates the heading port in a ship's voyage. This information is widely referred to by port operators for traffic estimation, and by shipping traders for supply forecasting, etc. However, we find that a considerable proportion (nearly 40%) of this information has been erroneously entered, both intentionally and unintentionally. In this paper, we aim to propose targeted policies to correct the inaccurate reports based on assessing the probability of observing wrong destination port reports of ships in AIS. To this end, we first of all conduct extensive interviews with relevant shipping stakeholders to understand the reasons behind the wrong destination reports. Second, based on the interviews and relevant literature we propose the influence factors. Third, we generate a data sample set based on the voyages performed by Capesize and Panamax bulk ships around the globe in a year. To generate this sample set, we leverage data mining techniques to extract the information from an AIS database and other databases. Finally, a discrete choice model is built to achieve the proposed objective. The results demonstrate that our model has an 84.1% accuracy rate in ascertaining the correctness of destination reports observed in AIS. We also find that, for a voyage, the speed of the ship, the historical accuracy rate of destination reports made by the ship, and the distance between the recognized origin and the reported destination of the voyage, have the most significant impacts on the accuracy of the destination report. Based on the findings, we provide managerial and policy suggestions to ship operators, port authorities, and regulators. [ABSTRACT FROM AUTHOR]

Improving transport connectivity has become increasingly significant to international trade research and transport policy agendas. Little has been done so far to investigate the direct association between maritime connectivity and economic growth, let alone the exact effect magnitude. As the connection between ports, local economies, and global logistics networks becomes ever closer, the economic role of port connectivity deserves closer attention. This paper uses the Panama Canal expansion as a quasi-natural experiment, and proposes a dynamic difference-in-differences (DID) method for making causal inferences about the impact of port connectivity on regional economic growth. This study leverages data mining of multiple sources of satellite data to provide innovative measures of ship movement, port area, and economic activity. We construct an index based on the global shipping network data from the Automatic Identification System (AIS) to measure the extent to which the ports are exposed to the canal expansion. This allows careful handling of the endogeneity problem. Change in the port area is used as a proxy for port capacity expansion, and is identified via satellite imaging. Night light data provide a measure of regional economic activity around the ports, and of the effect of port connectivity on regional economic growth. Our findings suggest that improvements in port connectivity have a significant and fairly durable stimulating effect on regional economic growth, and that port infrastructure expansion is one potential channel for realizing this effect. [ABSTRACT FROM AUTHOR]

The recent experience of lockdowns during COVID-19 highlights the prolonged impact a pandemic could have on ports and the shipping industry. This paper uses port call data derived from the Automatic Identification System (AIS) reports from the world's 30 largest container ports to quantify both the immediate and longer-term impact of national COVID-19 lockdown policies on global shipping flows. The analysis uses the Difference-in-Difference (DID) and combined regression discontinuity design (RDD)-DID models to represent the effects of lockdown policies. The combination of RDD and DID models is particularly effective because it can mitigate time trends in the data, e.g., the Chinese New Year effect on Chinese ports. This study further examines the potential shock propagation effects, namely, how lockdown policy in one country (i.e., China) can affect the number of port calls in other countries. We categorize ports in other countries into a high-connectivity (with Chinese ports) group and a low-connectivity group, using a proposed connectivity index with China derived from individual vessel trajectories obtained from the AIS data. The results provide a clearly measurable picture of the kinds of trade shocks and consequent pattern changes in port calls over time caused by responses to lockdown policies of varying levels of stringency. We further document the existence of significant shock propagation effects. As the risk of pandemics rises in the twenty-first century, these results can be used by policy makers to assess the potential impact of different levels of lockdown policy on the maritime industry and trade flows more broadly. Maritime players can also use findings such as these to manage their capacity during lockdowns more effectively and to respond more flexibly to changing demand in seaborne transportation. [ABSTRACT FROM AUTHOR]

Seaports play an important role in the global shipping network. Shipping participants often attach great importance to the measurement of container port connectivity, as it reflects countries' access to world markets. As a result, various port connectivity index systems have been proposed by members of the shipping industry and scholars. In recent years, technological developments especially the advancement of high coverage and real-time Automatic Identification System (AIS) data, have provided a chance to improve the scope and frequency of the existing index systems. An improved system is expected to reflect the dynamic changes in a port's connectivity which may be induced by either local disruptions or shocks in the wider economy. This study builds a monthly container port connectivity index system by applying big data mining techniques, graph theory, and principal component analysis (PCA) to AIS data, taking both port factors and shipping network factors into consideration. AIS records from 2020 are used to calculate the connectivity score of 25 major container ports. We also compare our system with the connectivity index commonly used in the shipping industry, the Liner Shipping Connectivity Index (LSCI). Our results show that the measurement of connectivity can be improved over indices that depend primarily on indicators of traffic volume. Ports like Antwerp and Tanjung Pelepas rank high in the proposed system due to their sound performance on their accessibility and strategic position in the local region instead of their traffic volume. The monthly index system is also proven to reflect timely changes in the shipping industry through its accurate portrayal of changes in port connectivity during the COVID-19 outbreak. [ABSTRACT FROM AUTHOR]

Ship navigation information derived from the Automatic Identification System (AIS) is widely used in the shipping industry. As AIS reports are featured with huge volume and noises, preprocessing of AIS data is essential before its further application. This study aims to develop effective algorithms to denoise and compress raw trajectories derived from AIS reports. Specifically, an effective noise detection method based on statistical theory and sliding window is first proposed to identify glitches in a given trajectory. Linear interpolation is further used to rectify the glitches detected. Then, two modes of algorithms are proposed for trajectory compression: static mode with preset threshold for compression and dynamic mode considering the distance between trajectory points and the coastline in a real-time manner. Numerical experiments show that the noise detection and rectification algorithms and the trajectory compression algorithms are accurate and highly efficient considering compression rate, information loss, and computation time. The main innovation of this research includes developing an accurate and robust trajectory glitch detection and rectification algorithm, proposing two modes of trajectory compression algorithms, and combining the two tasks in a holistic robust framework. Especially, the dynamic compression mode can overcome a major problem encountered in static compression where the compressed trajectories may go across land. It can also deal with more flexible compression requirements and thus should be more applicable in practice. • Developed a noise detection algorithm based on statistical theory to identify glitches in a trajectory from raw AIS data. • Used linear interpolation method to rectify the glitches detected in a trajectory. • Proposed effective online trajectory compression algorithm in static and dynamic mode (considering point-coastline distance). • Combined noise detection and rectification with trajectory compression algorithm in a holistic and robust framework. • Evaluated trajectory compression performance using various metrics under varied scenarios. [ABSTRACT FROM AUTHOR]

• Empirically explore the factors that affect shipowners' energy choices. • Apply various data mining methods and a threshold discrete choice model. • Address the heterogeneity of the carriers towards environmental awareness. • Provide important practical implications for responding to the new emissions regulations. On January 1st, 2020, the International Maritime Organization (IMO) implemented a new regulation for a 0.50% global sulphur cap for marine fuels, which was a dramatic decrease from the previous emissions cap of 3.5%. The new regulation will have an enormous impact on the shipping market. At present, there are different feasible schemes for reducing sulphur emissions from ships. Shipowners need to consider the economic cost, energy feasibility, and other relevant factors of different schemes before making decisions. This paper empirically explores the factors that affect shipowners' energy choices. Based on the new emerging individual ship dynamic data, Automatic Identification System (AIS), and other relevant databases, we apply various data mining methods and a threshold discrete choice model combined with an oversampling technique to conduct quantitative measurements and statistical analyses of factors for each ship type that affect the shipowners' choices. Three groups of indicators, including ship characteristics, shipowner characteristics, and market conditions, are considered in our analysis. In the model, we also address the heterogeneity of the carriers towards environmental awareness. This study provides important practical implications for responding to the new emissions regulations among maritime and maritime-related industries and policymakers. [ABSTRACT FROM AUTHOR]

Vessel trajectory prediction is a critical aspect of ensuring maritime traffic safety and avoiding collisions. The long short-term memory (LSTM) network and its extensions have represented powerful ability of vessel trajectory prediction. However, the previous studies often did not take dynamic interactions between neighboring vessels into account. Additionally, in complex traffic conditions, trajectory prediction will acquire uncertainty, and these potential negative factors can limit the prediction of future trajectory. To enhance the prediction performance, we propose an interactive vessel trajectory prediction framework (i.e., QSD-LSTM) based on LSTM, which is embedded with the quaternion ship domain (QSD). The QSD is beneficial for avoiding unwanted collision between neighboring vessels. In addition, the operation of trajectory clustering is further incorporated into our trajectory prediction framework, potentially leading to more robust prediction results. Numerous experiments have been implemented on realistic automatic identification system (AIS)-based vessel trajectories to compare our QSD-LSTM with several state-of-the-art prediction methods. The prediction results have demonstrated the superior performance of our method in terms of both quantitative and qualitative evaluations. [ABSTRACT FROM AUTHOR]