Digging Deep for Data on Safety: A Text Analysis Tool for Safety Planning on Construction Sites

Impact: Health and wellbeing, Technological


Despite decades of advances in health and safety, the construction industry sees many workplace accidents. Construction makes up nearly a quarter of the 123 workers who lost their lives in work-related accidents across all industries in the UK in the 12 months to March 2022. That’s on top of 81,000 people living with ill-health suffered at work. In the quest to prevent people from being harmed at work, the UK Health and Safety Executive (HSE) has initiated the platform for constructors to submit accident reports under the Reporting of Injuries, Diseases and Dangerous Occurrences Regulations (RIDDOR) since 2013. The information gathered from RIDDOR reports and other safety documents is vital for risk assessment for the industry—however, trawling through the vast volume of information generated by accidents has been a slow, laborious task. A text search system is well needed for users to effectively access and understand these data.

The Manchester research team led by Professor Sophia Ananiadou at National Centre for Text Mining (NaCTeM), in collaboration with the HSE, has developed a RIDDOR Text Analysis Tool to boost safety planning on construction sites. This Text Analysis Tool is a critical element of the Discovering Safety Programme, an ambitious scientific endeavour funded by the Lloyd’s Register Foundation. It uses natural language processing and machine learning to perform semantic searching on accident and incident textual data. By training computers how to read text almost like a human, NaCTeM’s experts have created a system that can do in-depth analysis instantly.

The tool is a big step forward. It ‘mines’ through the free text of the HSE documents to explore the contents and present the relevant information in an instantly usable form. This will enable health and safety managers, contractors and HSE inspectors to extract pertinent safety-critical concepts and associations without the labour of trawling through thousands of pages of text. It brings together HSE data and the power of artificial intelligence—in particular natural language processing and deep learning—to make devising risk assessments more accurate, effective, intuitive, interactive and much easier.

The system behind the RIDDOR Text Analysis Tool can be tailored for use in any industry. By proving that applying Natural Language Processing (NLP) tools to safety data works efficiently and effectively—performance of the backend NLP tools has accuracy of up to 90% compared to traditional keyword approaches—the tool opens the door to new, more efficient searching of valuable health and safety information for all kinds of industries. The next steps will include engaging with construction companies to develop the system further, then taking it wider, to other industries.
Category of impactHealth and wellbeing, Technological
Impact levelAdoption

Research Beacons, Institutes and Platforms

  • Biotechnology
  • Digital Futures
  • Institute for Data Science and AI
  • Manchester Institute of Biotechnology