At SedimentIQ, we have built an affordable, automated analytics solution to improve pit fleet operations at small-to-medium sized mines. We took a look at the landscape of mining analytics solutions and recognized two major problems. Existing offerings have very high implementation costs and produce a torrent of raw data, leaving supervisors struggling to make sense of it all.

SedimentIQ reduces implementation costs to near-zero using out-of-the-box setup and off-the-shelf hardware. Here's how it works: SedimentIQ ships you a box with Android phones and mounts; then, one of your on-site engineers plugs these phones in to power on your drills, loaders, and trucks. These phones automatically collect and feed data into our automated insights platform, which provides targeted analytics for your site using machine learning.

At an open pit mine in Southern US, our system accomplished the following within a day:

  • Automatically tracked machine utilization hours, saving hours on manual data collection ⌛
  • Accurately measured haul cycle times, equipping managers to improve hauling operations 📈
  • Highlighted queuing during haul cycles, finding opportunities to reduce fuel costs 💵

If you are interested in learning more about how SedimentIQ can be setup in an hour to improve pit fleet operations, continue reading our case study on a mine in the Southern US!

Note: While the following setup is for an open-pit mine, our product also works for underground mines where we use bluetooth beacons (5+ year battery life) to create a network of zone-based location without any connectivity!

Setup

We shipped an Android phone pre-configured with the SedimentIQ mobile app to the mine site. An on-site engineer connected the phone to power via 12 V outlet in a haul truck and turned it on; that's it! At this site, we're using cellular data to transmit data to our servers; if your site has no networking infrastructure (or if your site is underground), we'll send you another phone to collect data over Bluetooth.

The SedimentIQ mobile app collects GPS, accelerometer, gyroscope, and Bluetooth data automatically throughout the shifts. In the case of underground mines, we also ship battery-powered Bluetooth beacons to provide zone-based location in the mine. For the sake of confidentiality, we've removed any identifying information from this data and shifted timestamps.

https://s3-us-west-2.amazonaws.com/secure.notion-static.com/91a7eb82-0353-4ab9-bbcb-fb09440c9339/Untitled.png
Left: GPS data from 1 day; Right: one-minute slice of raw accelerometer data from the sample day

Automatically measure truck cycles

Whenever the smartphone gets access to a cellular data signal, it automatically uploads data into the SedimentIQ insights engine. One of the biggest points of feedback we've heard from the industry is that providing data isn't enough, so we've built an insights engine that turns this data into targeted insights for small to medium sized mining operations.

We use pattern recognition and machine learning techniques to measure and analyze cycle times, showing you opportunities to improve operations. One of the patterns the insights engine looks for is for progression through a sequence of geo-fenced areas as you can see below.

https://s3-us-west-2.amazonaws.com/secure.notion-static.com/7a645c17-eded-4dca-b115-d847cff9baab/Untitled.png
Left: Geofenced areas at the mine, Right: Automatically measured haul cycle times

When a truck travels from P1 to P5 and back to P1, the insights engine automatically identifies this as a hauling cycle and will display relevant metrics related to this cycle to identify areas for improvement in mining operations. For the above example, the system identified:

  • 32 haul cycles on the sample day. Given, the truck has a 50 ton hauling capacity, we augment that information to track the total tonnage (in this case: 1600 tonnes)
  • 16:54 min average time of the haul cycles. We augment this information with worker schedule to identify equipment and worker variability in a given time period
  • Queuing and idle times (more on that below)

The SedimentIQ insights engine can also be customized to automatically track cycle times for many different tasks. For instance, if your mine or quarry generates waste rock which is hauled to a separate stockpile, the system will create a new sequence to track waste haulage cycles as well. For underground operations, we use Bluetooth beacons to provide zone-based location tracking instead of GPS geofences. The SedimentIQ dashboard allows managers and engineers to easily create new tasks and analyze performance.

https://s3-us-west-2.amazonaws.com/secure.notion-static.com/308aba49-f9fa-4122-8683-4386499c2657/Untitled.png
The SedimentIQ dashboard allows you to customize tasks and view cycle analytics

Automatically measure queuing

The SedimentIQ mobile app also collects data from other smartphone sensors such as the accelerometer and gyroscope. These sensors, typically used to figure out when users turn their phone, allow us to detect vibrations produced by a running truck or loader. Using machine learning techniques, the SedimentIQ mobile app produces an "activity score" from raw data that can be used to determine whether a machine is parked, idling, or performing productive work.

https://s3-us-west-2.amazonaws.com/secure.notion-static.com/c0081565-25f9-46ba-829d-d36d1b1922c1/Untitled.png
For e.g., we use a threshold heuristic (score=0.6) to determine if a machine is parked or idling

The insights engine combines machine status with positional information to determine whether the machine was queueing during a given cycle. For example, if the machine was idling near the plant, the insights engine automatically categorizes this as a queuing delay in the haul cycle. On the other hand, if the machine was off and parked, the insights engine attributes this to shift breaks, such as lunch. On the sample day, we can see that one of the longer cycles early in the data was caused by an 11-minute queuing period at the processing plant:

https://s3-us-west-2.amazonaws.com/secure.notion-static.com/62d03883-e879-45eb-be75-50d2f6ef00dd/Untitled.png

A similar approach can be used in underground operations, where machine status is combined with Bluetooth beacon proximity to categorize delays. Using this technique, the insights engine can help your supervisors and engineers dig into repeated delays and optimize operations.

Takeaways

In this post, we took a tour through how a smartphone running the SedimentIQ mobile app can be set up to track cycle times and delays in under an hour. The mobile app automatically sends data to our insights engine whenever it has a network connection. Lastly, we examined how the SedimentIQ insights engine can provide targeted, customized analytics to improve operations at your site.

Interested in trying something similar at your mine or quarry? Have any feedback or questions? We'd love to hear from you and ship you a pre-configured mobile phone for a free trial! Please reach out to us at [email protected].