Image source: Photo by Shane McLendon on Unsplash

In order to identify workplace risk, hundreds of safety checklists are collated and processed daily at any given Anglo American mine. If these checklists are not processed quickly enough, risks could go unnoticed, resulting in potentially life-threatening situations. Automating these processes is a critical way of addressing safety, with far-reaching benefits.

In its ongoing effort to improve both safety and efficiency with innovative practices and technologies, Anglo American sought to enlist the help of an experienced machine learning company, and DotModus was well-placed to assist.

Using Google Cloud, in concert with digital solutions specialists Integrove, DotModus has delivered a machine-learning powered document-scanning solution that is less vulnerable to backlogs, delays and human error, and is not reliant on the availability of tablets or other electronic equipment in the field. The automated solution brings attention to risks sooner, streamlines the work allocation process, and assists Anglo American in keeping their mine workers safe.

The successful implementation of the application at multiple Anglo American mines across South Africa prompted a worldwide rollout to 42 other sites globally. The application has subsequently been translated into several languages and is being used at Anglo American and De Beers mines in Botswana, Namibia, Zimbabwe, Chile, Brazil, Peru, Canada, and Australia.

The solution makes use of custom processing and Google Cloud’s Vision API to extract both handwritten and printed text from safety checklists that have been scanned in bulk.

There are more than 60,000 permutations of checklists that can be completed and scanned into the application. Therefore, to speed up the extraction process, the system is pretrained on blank versions of all possible checklists. As each checklist passes through, the system identifies:

  • the type of checklist (via QR code)
  • its unique layout
  • how it has been filled in and
  • which extraction process to trigger
  • re-alignment adjustments to correct tilted or skewed pages
  • contrast and lightness adjustments to better extract data

The extracted data is then processed, collated and uploaded to Anglo American’s broader cloud environment for use during work allocation sessions.

The power and efficiency of the Google Cloud Platform is particularly evident on this project. A single instance serves Anglo American’s global needs, and scales to usage in any time zone and geography. The result is a solution that is fast, reliable, scalable, highly available, and easy to monitor and manage.

The design and implementation of this human-centric technology by Anglo and DotModus provides an outstanding example of how technology can be used to change and improve people’s lives.