How a Chilean Copper Mine Used an AI-Based Solution to Prevent 144 Hours of Lost Production

2019

Bahram Sameti, Motion Metrics International Corp.; Saeed Karimifard, Motion Metrics International; Caitlin McKinnon, Motion Metrics International Corp.; Jose Oliva, Motion Metrics Latin America S.p.A.; Shahram Tafazoli, Motion Metrics International Corp.; Derek Cooper, Motion Metrics International Corp

Loader bucket teeth are prone to breaking off during operation.  These missing teeth can go undetected by equipment operators or be transported downstream in the excavated ore where they can obstruct crushers.  Clearing an obstructed crusher is a highly dangerous procedure that requires several hours of crusher downtime and results in significant financial losses.  Like many hard-rock mines worldwide, the Chilean copper mine in this case study experienced significant production loss from tooth breakage – between 2012 and 2015, the mine attributed 153 hours of crusher downtime to obstructions. The mine proactively mitigated the impact from missing bucket teeth by installing a missing tooth detection system fleetwide, including on two loaders.  This computer vision-based solution uses a ruggedized thermal camera and deep learning algorithms to continually monitor the loader bucket; when a missing tooth is detected, the machine operator is alerted via audio-visual alarms from an in-cab monitor.  Since the AI-based solution detects broken teeth in near real-time, missing teeth can be intercepted before they travel downstream. During the last 10 months, the system has detected 12 missing loader teeth.  Since installation, the mine has experienced zero crusher downtime due to missing loader teeth and, according to conservative estimates, avoided a production loss of approximately 144 hours because of the missing tooth detection solution.  Moreover, the mine has significantly improved mine safety by preventing dangerous crusher jams.  This case study will: (a) provide context for the missing tooth problem at the Chilean copper mine; (b) explain how the installed detection solution works; and (c) estimate the production losses prevented because of the mine’s risk mitigation strategy.
$0.00