Motion Metrics: Missing Shovel Teeth Detection – Product Review

Posted by on May 27, 2014 in Favorites | 0 comments

Motion Metrics: Missing Shovel Teeth Detection – Product Review

I started this blog to provide a platform for discussion on operational excellence with a focus on the utilization of technology. As the blog has evolved based on the readership response such as my post here on career options as a mining engineer, I’ve decided to try something new today – a detailed product review on a mine applications technology: ShovelMetricsTM, a product designed & manufactured by Motion Metrics International Corp. This will be longer than my usual blogposts so I hope you enjoy it (alternatively, you can scroll to the bottom for the TL;DR conclusion)

With the goal of enhancing mine productivity and safety for the past 14 years, Motion Metrics has a suite of five (5) products but here in this review, we’ll just be focusing on ShovelMetricsTM and its image-based tooth monitoring technology.

What is it?

ShovelMetrics is a monitoring system for mining shovels. It is well-known for its missing tooth detection (Previously known as ToothMetrics) and tooth wear monitoring capability. Shovel teeth can break off during operation and once lost, they go into either the ore or waste steam, causing downstream issues . If you receive an alert when a tooth breaks off and ends up in your haul truck that is driving to the crusher, you can stop it before the tooth likely ends up damaging and/or plugging up the crusher. If the metal does pass the crusher, it could cause even more damage on the conveyor belts such as tearing the belt. In addition,  ShovelMetrics has a suite of other features for shovels including proximity detection, fragmentation analysis and payload monitoring.

The installation of the ShovelMetricsTM tooth monitoring system includes a CPU, power supply, bucket camera/bracket & LED light, and an in-cab monitor with armored cables running throughout the shovel. With a live screen of teeth status (see photo), the camera is able to translate photos into usable data of all of the bucket’s teeth and its current status: good, warning or alarm. In addition to alerting the mine when there is a missing tooth, the product will also track and report trends in tooth wear. This data can be used to create a tooth change-out schedule to avoid longer fill times due to reduced digging efficiency and to broken teeth leading to possible crusher downtime and an unplanned tooth change-out.

For proximity detection, radar proximity sensors and surveillance cameras can be added to the system. The system can also perform fragmentation analysis of the material in the shovel bucket using the same camera that is used to monitor the teeth. Lastly, for hydraulic shovels and excavators, the ShovelMetrics system can also provide bucket-by-bucket payload monitoring. While these three additional features have easy ROI’s and are great, there are a lot of other competing products that offer similar results and challenges (such as Split-ShovelCam for fragmentation analysis and Caterpillar’s TPMS for payload monitoring).

Motion Metrics would likely argue that their products have advantages over other systems such as the 2.5% accuracy that their payload system provides in a controlled test environment. However, I personally don’t believe actual performance and its variance at a mine site’s operations level is within a 2.5% margin of error. Another example is their proximity detection solution, which faces the same challenges as most proximity warning systems where the effectiveness decreases as nuisance alarms increase. (I did my Bachelor’s graduate paper on Collision Avoidance Systems and most of the challenges back in 2005 are still the same today unfortunately).

To me and to those operators that I’ve talked to, the biggest value proposition of Motion Metrics is their proprietary and successful algorithm on detecting worn or lost shovel teeth. However, if you were going to install ShovelMetricsTM anyway, it would be easy to single source all the other value added products directly from Motion Metrics.

Business Case

So you might ask: “Where is the Return on Investment here?” so we need to answer “How much does it cost when a crusher goes down?”

Let’s use a generic open pit gold mine example.

Total number of crushers: 3

Crusher design capacity: 40ktpd each, 120ktpd total

Design Utilization: 80%

Max Capacity: 50ktpd each, 150ktpd total

Ore Grade: 0.05oz/ton

Au Price (as of May 11, 2014): USD$1290/troy oz

If one crusher goes down for 4 hours due to a shovel tooth plug, then you lose 6,667 tons into that crusher. Theoretically, you could divert a half of the loss tonnages to the two remaining crushers and run them at full capacity for 4 hours. But you are still stuck with a loss of 3,333 tons for the day. Using the assumptions above, it results in a delay of $214,978 (3,333tons x 0.05oz/ton x $1290/oz) in revenue. And as we know, mining economics is all about a dollar today is worth more than a dollar tomorrow.

As it is confidential information, I cannot release the cost of ShovelMetricsTM but as one miner said to me, “Avoiding one crusher shutdown pays for it” and as I saw in another mining company’s newsletter, “Assuming a 10 hour production delay, the cost of the entire ToothMetrics system is offset if a single tooth is prevented from entering the crusher.”

Do note that the real cost of crusher downtime is reduced if your mine to mill process isn’t optimized (i.e. crusher utilization issues resulting in increased truck queue time).

There are a number of other factors that I did not take into account for this simple calculation, such as hedged gold contracts, mine to mill process is not optimized so you weren’t running the crusher at design capacity anyway, other crusher utilization issues, cost of power, NPV, IRR and a company’s internal hurdle, blending implications. The list goes on! If you find mining economics interesting, the Colorado School of Mines offers a short course. Details here.

Business case and productivity aside, there is a huge safety component to not sending metal to a crusher. As cheesy as the music may be, this YouTube video posted here shows a horrific safety incident when a worker is working on a piece of jammed steel in the crusher.

The Future: Wearable Tech

You’ve all heard of the “wearable tech” craze (Fit Bit, Pebble, GoPro Hero3+; even Amazon has a dedicated Wearable Technology storefront that was launched in April 2014!). And believe it or not, it’s coming to mining! These are some of my thoughts on wearable tech:

  1. It’ll be challenging to run privately and may cause security issues as most backhauls data over low power Bluetooth (BLE), an insecure protocol so equivalent to broadcasting in the open. Anyone can discover which services you’re exposing.
  2. It’ll be tough for applications that are always on, as most is powered by coin cell battery. You’ll either need to recharge frequently or if you let the batteries completely die, they are tough to replace.
  3. They are not currently designed for ruggedized environment.
  4. It needs to backhaul to a cell phone or a computer so you’ll still need to have one or the other nearby for it to be useful. The wearable tech can store data but you’ll still need to download it at some point (limited network infrastructure at either an underground or open pit mine).
  5. You cannot host an application natively on most as there is only an API option for app development for google glass until recently (SDK development pending). In that case, applications are limited and everything is server backed and there could be latency issues (in mining, many applications require low latency). Therefore, it is designed for content consumption and not for interaction.
  6. The obvious safety issues – situational awareness, distraction, peripheral vision block.

Motion Metrics teamed up with Vandrico and showcased what they believe is the future: wearable technology for mining, which sends alerts to the mine operators. They call this technology MetricsGear. See the photo of us trying out a pair of Google Glass synced up to the alerts of BeltMetrics on display during the CIM convention in Vancouver (May 2014). They also have their system connected to a smart watch that vibrates when there’s an alert and displays relevant logs. Challenges aside, I love the innovation and “outside the box” approach here!


I have identified five challenges/opportunities in deploying & using the ShovelMetrics system. They are as follows:

1. Downstream process to act on information provided by technology.

This is a challenge not only for shovel teeth detection products but also for all mine applications technology out there. I say this over and over again (old blogpost here) – all this data provides a report, which then is supposed to generate a downstream action item for mine operations. Then and only then will you get payback (ROI) for your new shiny toy. ShovelMetricsTM will only be able to tell you when you need to change out your tooth or when a tooth is missing; mine operations & mine maintenance will need to work together on scheduling out the work.

Currently, ShovelMetrics does not log every bucket, which was one of the gripes that a user had with having a comprehensive data set for downstream processing. According to Motion Metrics, “The frequency of missing tooth logs is largely dependent on the quality of the camera calibration and environmental factors. With a properly calibrated system, the system can typically log an image every one or two loading passes.”

2. Single unified hardware CPU hub & display

As much as OEMs and OTMs are fighting it, the industry is moving towards a unified hardware hub & display (see work done by GMSG’s working group here). From a technology advisor for a major mining company: “I think ToothMetrics has value for hard rock mines – but doesn’t need a dedicated screen. The concept is sound, but just posting a warning through FMS [fleet management system] would be as effective.”

As Motion Metrics is a participant and contributor of the GMSG Situational Awareness working group, it should be natural for them to start opening up commercial agreements to work on a 3rd party hardware CPU or display. However, a prospective customer of Motion Metrics told me that discussions around a single hub & display were “non-committal” in late 2013. My hope is that Motion Metrics sees both the value and commercial demand of utilizing 3rd party hardware and moves toward productization.

3. Integration with fleet management system – overall industry issue

Currently, their payload monitoring component integrates with Wenco & Modular Mining dispatch systems. However, it does not integrate with others such as Micromine, Leica and Cat systems. Nor does ShovelMetrics integrate into FMS as a total solution. This is an overall industry in mining where every single vendor wants to have as much of the pie as possible. Integrations and collaboration are only happening at a very high level such as at the GMSG working group or happens when there is a huge push of customer driven pressure.

I get it. I’ve read the Apple philosophy that they want to control the entire end-to-end customer experience so they won’t license their O/S out to 3rd party hardware. But these aren’t operating systems! These are applications. Most of us have heard the analogy with an iPhone before. We have multiple applications such as Facebook, Instagram and Google Maps on the same hardware CPU, display and all utilizing the same GPS unit. Wouldn’t it be nice if we could move together towards a ruggedized version for mining?

4. When the system is down

As much as I love technology, there is something to be said about knowing how to do things the old school way (such as a surveyor staking out high grade after a blast instead of relying on the HPGPS on board to tell you). When your technology goes out (broken hub on the equipment, network infrastructure and communications is down), a mine generally does not stop operating! So if you are putting in all of these technologies, don’t forget to build or update your Standard Operating Procedures (SOP) and also train everyone on what to do when technology disappoints.

One company tries to mitigate this problem by using Motion Metrics as a “first line of defense” and they are evaluating a secondary solution with an embedded RFID tag inside the hollow teeth as an extra layer of redundancy.

5. Oil Sands Application

During CIM in Vancouver, I also had a quick discussion with Chris Langmead (Manager, Mine Systems & Reporting from Suncor Energy, Oilsands). He believes that ShovelMetrics was previously tested in the oil sands operating environment; however, due to the dark nature of the tar sands, it “fooled” the ShovelMetrics camera into thinking that the tooth was gone. E-mail request sent to CNRL, a current ShovelMetrics user in the oil sands, for current product performance was not returned. Feedback from Motion Metrics, “The oil sands operating environment does pose new challenges for our system. However, we are actively working with our customers to improve the performance of the system. One of the key reasons why we developed the new ShovelMetrics CPU platform was to improve the system performance under all environments and conditions.”

TL;DR I want to thank all the mining companies that contributed to the article. So in conclusion, I truly believe in the direction of Motion Metrics and I do like their flat level reporting so there is less time spent on bureaucracy and more on innovation. If you’ve ever had shovel teeth breakage and downstream impact, this is a no brainer ROI. In 14 years, they have achieved over 230 installations globally at 50+ mines. But I also think that they have a huge opportunity to break out of the conventional revenue driven strategy, and into a more integrative approach within the mining industry.

For more information on Motion Metrics, contact Enoch Chow at



Compensation was provided by Motion Metrics International Corp. The opinions expressed herein are those of the author and are not indicative of the opinions or positions of Motion Metrics.