Challenge

A large mining company approached Mosaic Software seeking help to prove the feasibility of approaches for solving two problems. The first problem was to map water bodies covered with ice and snow accurately and to determine when mining equipment approached these hidden bodies of water. Solving this problem could help prevent operators of large mining equipment from driving over these water bodies and sinking into them. The second problem was to identify cases where bulldozer operators violated a safety boundary behind massive mining shovels. These violations created a risk that the dozers could be crushed by the 200-ton counterweight on the back of these shovels. The goal for both of these problems was to increase the safety and productivity of mining operations.


Image from the hidden water hazard
Image from the equipment proximity hazard

Action

To avoid harm to individuals and losses in revenue and time with the possibility of these two hazards, this company hired Mosaic Software because of our expertise with geospatial data analysis and software development. Mosaic Software used the company’s Microsoft Azure environment to analyze the Global Positioning System (GPS) data and to create an Azure Maps display. The GPS data was coupled with satellite data from shapefiles and overlaid on the Azure Maps display to highlight the historical proximity of mining equipment to hazardous locations. Mosaic Software added playback capability and video overlays to allow supervisors and managers to easily review dozer movements so they could understand the full context of dangerous operations.

Part of this work involved analyzing 44M rows of high-precision, high-frequency data representing the positions of dozers and shovels for a complete month. The goal of the analysis was to identify boundary incursions between dozers and shovels for reporting and visualization. Mosaic Software developed a carefully optimized custom methodology for processing this data, which allowed the full dataset to be analyzed in four minutes, where non-optimized approaches had previously consumed 20 hours. This rapid processing allowed Mosaic Software to report quickly on the analysis approach, presenting updated results to the customer regularly and incorporating feedback into successive analyses. The final results were far superior to what they could have been because of this optimized approach

Results

Mosaic Software was able to demonstrate the feasibility of using GPS and satellite data to detect and mitigate risks due to unseen hazards. Working with the customer, we determined that satellite data accurately represented the water body locations and that GPS data was sufficiently accurate to detect dangerous proximity between dozers and hazards. Having proven the feasibility of these approaches, the customer was then able to incorporate them into their systems and processes to increase the safety and profitability of their mining operations.