Maritime Accidents: Root Causes and Saving Lives

 

The Approach

  • Investigating the current reporting methods and documentation used by CHIRP, particularly in light of the improvements made to their reporting system last year. 
  • By means of statistical sampling techniques we were able to select a representative sample for text analysis. The text analysis revealed themes in line with previous research that identifies ‘deadly dozen factors.’ 
  • The analysis outlines a series of recommendation for improving data quality going forward. 
  • Topic Analysis based on Latent Dirichlet Allocation revealed different themes according to the severity of the accident. 
  • A proof-of-concept predictive model was developed to impute the field of incident severity. To build the model, we used feature engineering techniques that transformed verbatim data into vectors, so the data is on a format amenable for Machine Learning models. 

The Client

CHIRP is an independent and impartial charity dedicated to improving safety in the air and at sea. Their confidential human factors incident reporting system empowers people working in the maritime and aviation sectors to share their safety concerns without the fear of being identified. 

The Client's Problem

CHIRP wanted help to leverage the information collected about maritime incidents to draw insights from the data to improve their reporting system going forward. 

The Solution

  • Text Analysis of key themes around report severity.
  • Recommendation to streamline and improve quality of reporting going forward. 
  • Predictive Model to impute missing field of report severity.

The Benefits

  • Suggestions to improve the reporting system. 
  • Identifying primary causes of accidents and pattern recognition.
  • Key Insights about drivers of accident severity.
  • Missing data imputation methodology that can be developed further. 

Not only did the project exceed expectations for CHIRP, but several options for moving forward were also identified. The project helped CHIRP learn a lot about the data that they hold and outlined a future approach which will make the collection and analysis of future data more meaningful. For CHIRP “the project was more than worthwhile!” 

Other Case Studies