AIS Data Profiling For Error Analysis In Indian Ocean Region (IOR)


The Project discusses the importance of the Automatic Identification System (AIS) in enhancing safety and efficiency in maritime navigation. It highlights the potential negative impact of fast mandatory implementation of AIS without adequate research on its use, citing examples of AIS-assisted collisions. The need for specific dominant errors in AIS to be identified and remedial actions suggested is also highlighted as a way of preventing accidents.

It also provides a summary of previous studies conducted on AIS, including a VTS-based AIS study and a data-mining study. Specific domains critical to the work are also identified, including familiarity with Python modules and open-source tools for web-scraping and data analysis. The challenges and opportunities associated with the project, such as real-time analysis of data, writing complex Python scripts, and data cleansing, are also discussed.

Key highlights
  • The implementation of Automatic Identification System (AIS) aims to enhance safety and efficiency of navigation, safety of life at sea, and maritime environmental protection.
  • Previous studies have been conducted on AIS errors in the UK and Singapore Strait but not specifically for the Indian Ocean Region.
  • The project involves real-time analysis of the data, which can be a challenge due to the frequent updates of AIS data.
  • The use of Python modules like NumPy, Pandas, Matplotlib, and Seaborn can aid in visualizing and analyzing AIS data.
  • The lack of a reliable shipping registry for the Indian Ocean Region poses a challenge in finding a suitable source of data for the study.
  • Proper data cleansing is necessary before using AIS data for visualization and analysis.
Key Challenges
  • Real-time analysis of AIS data due to frequent updates.
  • Complexity in writing Python script for analyzing multiple AIS fields.
  • Lack of a reliable shipping registry for Indian Ocean Region.
  • Uncleansed data with several “Unnamed” columns, requiring proper data cleansing.
  • Analysis of seven fields and their errors, with data validation through web scraping leading to time and space complexity.
Major opportunities
  • Optimization of algorithms to handle the time and space complexity of the model.
  • Development of faster algorithms for analyzing AIS data.
  • Implementation of a distributed computing architecture to handle large volumes of data in real-time.
  • Collaboration with other shipping registries to improve data availability and reliability.
  • Improving data cleansing techniques to streamline the data analysis process.

“Transmission of erroneous information by AIS is an important issue that can affect its usefulness & Fast mandatory implementation of AIS equipment for SOLAS ships without adequate research on its use, may be having a negative impact on its success and hence endanger safety of marine navigation.”

Vatsal Maheshwari, Dr (Cdr) Arnab Das & Shridhar Prabhuraman