Can you cost-effectively track soap?
How we’re designing a low-cost electronic tracker for the bare essentials of humanitarian aid.
One of the first questions that I’m always asked when developing new technology is “How much will it cost?”. Often this question isn’t referring to the cost of developing a product. The big question is whether or not a product can be manufactured for a cost that will work for a business model. In the case of our pilot: Geoseals, we needed to discover whether it was possible to track aid internationally for a cost that was comparable to the value of the commodities being transported.
The aim of the Geoseals pilot is to develop a security seal, which when broken transmits its location. This is to determine the final delivery location of the package to enable aid distributors to make more informed decisions when designing and running aid distribution programmes.
Our visit to an FCDO aid warehouse in Sprint 1 showed us that a lot of aid commodities are often low value and shipped in high quantities. These commodities are really the bare essentials, the items anyone would need to survive: things like soap, a solar lamp, things to keep you warm and even just a pan to cook with. Imagine having to flee your home, the last thing on your mind is whether to pack a toothbrush or a saucepan. These kinds of commodities can cost anything from £1.50 up to £35.00 each. During our visit to the FCDO warehouse, we also saw higher-value items which are shipped in lower quantities such as long-term shelters and medical supplies.
Tracking technology can be expensive and in this sprint, we wanted to discover if it would be feasible to track these low-value, high-quantity commodities with GeoSeals. What we needed to do was to compare the cost of our proposed tracking technology with the cost of the aid commodities. This involved designing some systems and pricing them according to the electronic components required to build them.
The past 2 years have seen some of the worst component shortages in the electronics industry since records began with lead times doubling from 12 weeks to 22 weeks, with some components unavailable until 2023. This has pushed prices sky-high with manufacturers seeing unprecedented price increases as global stocks deplete. Predicting how much a product will cost in these market conditions is very difficult. However, the Arribada Initiative has been manufacturing tracking products during this tumultuous period and we’ve used this experience to help answer the big question.
“Will GeoSeals be low-cost enough to track low-value items at scale?”
What do you need to track a package?
Asset tracking technology is part of our lives, we track our Amazon packages, the bus we’re waiting for and even our pizza, from oven to door. These tracking technologies often rely on GPS and some way of identifying each asset, e.g. a barcode or RFID. For asset tracking to work the technology needs to track events. A tracking event is a timely event that records where the item was, at which time and what has happened to the item. For example, where has my package been delivered to or has my pizza left the oven yet? These tracking events are recorded and relayed to a central database which can be accessed by users. In the case of GeoSeals, being able to see where an aid commodity has been opened allows aid distributors to make informed decisions. The diagram below shows how a typical asset-tracking technology works.
GPS tracking technology is expensive and often trackers such as those used by delivery drivers rely on smartphones which have GPS built in. GeoSeals is expected to operate automatically without human interaction, it needs to:
- Detect when the package has been opened
- Gather data on where it is when this happens.
- Transmit the tracking event data back wirelessly wherever it is in the world.
The cost of implementing this kind of device would cost £150.00 per package which far exceeds the cost of the low-value items we’d like to track. We can’t rely on smartphones to provide the GPS position and this meant that we needed to take a different approach for GeoSeals. We began looking at ways to wirelessly detect multiple packages using a GeoSeals hub. The hub would track its location as well as detect and transmit the status of the surrounding packages to a database via cellular or satellite networks. This would significantly reduce the cost of tracking each item and potentially make it low-cost enough to be feasible.
Tracking Technologies
There are many different ways to track assets but two technologies stood out to us as being possible solutions to explore. BLE (Bluetooth Low Energy) Beacons and UHF (Ultra High Frequency) passive RFID (Radio Frequency Identification) tags.
BLE Beacon technology is widely available and is used for asset tracking and also for keeping track of personal items. Each item to be tracked would be fitted with a BLE beacon, these beacons would communicate with each other and relay their status back to a GeoSeals hub, which would record the location and transmit data back to the database. BLE beacons each have their own unique ID and they operate by scanning the area for any available BLE or Bluetooth devices, once a device has been detected and successfully paired data can be passed between the beacons. This method enables data to pass throughout the network by relaying it from beacon to beacon.
A BLE beacon network would enable GeoSeals to detect which packages are nearby, but this technology alone would not be able to detect when a package was opened and the BLE beacon would need to be fitted with additional sensors e.g. conductive ink or light sensors to detect this event. A BLE Beacon would cost around £20.00 to manufacture at 1000 off quantities, an example of a BLE Beacon is shown below.
RFID (Radio Frequency Identification) is widely used for asset tracking, contactless payments and as a shoplifting deterrent. You will have at some point experienced the alarms going off at the doors of a shop when a tag on an item hasn’t been removed. There are lots of different types of RFID that operate over different ranges, but we decided to focus on UHF (Ultra High Frequency) passive RFID (Radio Frequency Identification) tags. These tags don’t require any batteries and are widely used for asset tracking, they are available as a barcode sticker as shown below.
Passive tags are very inexpensive and are widely available, with each label costing approximately £0.20 each for a 1000 off quantities. The major cost in using this technology to track assets is the RFID transceiver. The RFID transceiver detects and communicates with tags, however, depending on the range required a single transceiver can detect any number of tags, with industrial readers scanning for over 1000 tags per second. An example of a handheld scanner capable of reading 1300 passive UHF RFID tags per second with a range of up to 6m is shown below.
Each item to be tracked would be fitted with two RFID labels, one of which would cover the opening panels of the packaging. This security RFID label would be placed such that it requires the label to be cut or broken to access the goods inside. RFID tags rely on an intact antenna to transceive data, once this is broken it can no longer be detected.
The secondary RFID label, the packaging tag, would be placed so that it is not broken or cut when a parcel is opened. This method would allow groups of packages to be scanned to detect not only the packages within a range but also whether the packages have been opened. If both RFID tags are detected, we can log that the package is still closed and intact. If there is no response from the RFID label sealing the box but the packaging tag is detected, this signals that the package has been opened as the seal tag is no longer responding. This simple solution could enable us to track packages at a very low cost due to the low cost of RFID antennas/tags.
For each group of packages to be tracked there would be a GeoSeals hub which would detect the security and packaging RFID tags on each box. This GeoSeals hub would also track its location and transmit the status of the surrounding packages to a database via cellular or satellite networks.
Smart Boxes
To create a network you first have to decide how far you’d like to transmit your data, this is known as the range and for GeoSeals we had to look at how commodities were packaged. Most of the low-value, high-quantity items were packaged in boxes which were then palletised on a standard-sized pallet. A standard-sized pallet is 1.2 x 1.0 x 1.2 m, based upon the commodities we’d seen on the FCDO warehouse visit we decided to assume that each pallet contained 42 items. We then compared the range of two technologies: BLE Beacons and UHF passive RFID tags, to see how many pallets and therefore how many items a single hub could detect.
The diagram above shows how the range varies depending on the type of technology used. What can be seen is that BLE Beacons have a larger range than UHF passive RFID however we found that they are significantly more expensive to implement. This finding showed us that the network range had a significant impact on total costs. The GeoSeals hub is the most expensive part of the system and if the system can only detect over a very short range this would mean more hubs would be needed per shipment.
This led us to consider how we could reduce the number of GeoSeals hubs by extending the network using the BLE Beacons. This would mean creating a hybrid network where each box was labelled with RFIDs and each pallet was fitted with a GeoSeals BLE beacon. The GeoSeals BLE beacon would detect which boxes were on its pallet and relay this information to the nearest GeoSeals BLE Beacon. Each shipment would have a pallet with a GeoSeals Hub which would receive the data from the BLE Beacons and transmit the data via mobile network or satellite to a database. This significantly increases the range of a GeoSeals tracking network so long as each BLE Beacon is within range any number of pallets could be tracked.
What does it cost?
We compiled the costs of manufacturing each part of the three systems we proposed. Below is the cost for each unique part of the system.
To truly get an idea of costs we needed to consider how many individual items needed to be tracked. We made an assumption that a typical shipment would contain 100 pallets each containing 42 boxes and that each item would be tracked. We used the estimated ranges to discover how many GeoSeal hubs and BLE Beacons would be required for the shipment. From this analysis, we found that a hybrid solution was the cheapest and it could potentially track an individual item for just under £1.00, below is a graph which shows the variation in commodity costs vs. the different system costs per tracked item.
As can be seen in the graph above, tracking individual packages using BLE beacons is the most expensive solution and is significantly more expensive than an RFID or Hybrid system. This research shows that GeoSeals would be viable for tracking low-cost and high-quantity aid commodities. It feels great to know this and whilst this is by no means a final costing it has given us a good indicator that GeoSeals would be cost-effective for tracking even the lowest cost commodities, those bare essential items which are so important to keeping people safe and well. It will give those people in charge, those making the important decisions, more information to help the people most in need.
This post has focussed on our findings on costs, but there were lots of other great things that came out of this sprint. There’s still plenty to do and as any engineer will know these figures are only estimates. There are some really important questions yet to answer around shipment size, supply chain and packaging types which will be essential to decide if the methods proposed here are truly feasible. We’re looking forward to testing our proposed systems and getting a better understanding of how we can build a cost-effective tracker for the bare essentials of humanitarian aid.
