A large building and construction group was seeking greater insights into serious incidents with Machine Learning on Google Cloud Platform.

The building and construction company sought to develop a greater understanding of the root causes from serious harm incidents. Data Science techniques using Machine Learning and text analysis were used to analyse existing data sources, in an effort to help identify trends across the various sites, situations and the company as a whole.

Our Solution

A Cloud Data and Analytics Environment was rapidly created onĀ  Google Cloud Platform (GCP). The Platform was designed using the following tools:

  • Google Cloud Storage
  • BigQuery
  • DataLab
  • Data Studio


Servian was able to quickly undertake topic analysis to both confirm and dispel a number of hypothesis correlated to near misses.

These insights have helped inform management about where to focus safety efforts to reduce harmful incidents.

From a Data Science capability perspective, the project demonstrated the value which could be extracted from existing data sources with machine learning.

Next Steps

The rapid time to market has enabled a quick discussion about the potential to expand the solution. The next steps include adding to the number of data sources and broadening the use of machine learning on GCP to help improve safety and other business use cases.