By: Stephen Lambe
With over 8 billion connected IoT devices and 2 billion GPS-equipped smartphones already online, logistics businesses are tracking assets in real-time at almost every step in the supply chain. The potential impact on operational efficiency, customer service, and on-the-job safety is estimated at 2 trillion dollars in economic value.
For example, Metromile, a usage-based car insurance provider, is using our platform to track where drivers park so they can help prevent parking violations and make it easier for users to find their car. Compology is using image-based sensors to help companies and cities track when dumpsters need to be emptied; they’re using our maps to contextualize all of this live data. Pole Star Global tracks marine vessels as they move through shipping lanes to ensure audit compliance. Pole Star is also using our tools to set up geofenced regions so operators know if they’re entering a potentially dangerous zone.
So what do all of these applications have in common? They share the four core components that comprise all asset-tracking solutions: location updates, stream processing, a database, and a map or dashboard. Let’s break these down.
1| Location updates
Location updates (containing an ID, latitude, longitude, and time stamp at a minimum) are what your sensors collect and send, providing a snapshot in time for each asset. A good example is a driver at a ride-sharing company using a driver app equipped with in-app navigation. When using the app, drivers transmit origin/destination information, route paths, or vehicle health data. All of these location updates get posted to an API gateway.
2| Stream Processor
From the API, the stream processor takes over. It receives the location updates and publishes real-time streaming data that you can access with your application stack. Further processing can be done with cloud functions to add special context to the asset updates. For example, you could call our Directions API to include distance from an asset to its destination at a given time. Popular stream processors include Apache Kafka, PubNub, and Amazon Kinesis. We recently held a webinar with PubNub on streaming and mapping data from IoT sensors.
3| Database
After being processed, your data heads to a NoSQL database for storage, where it can be quickly accessed using the front-end interface. Any special properties that are added with cloud functions (e.g. distance from destination) can be stored in a single field. The last step before displaying the data is converting the database JSON into a GeoJSON via an API gateway.
4| Map
The fourth and final component is the most visible and important — the tracking dashboard. We built the example below with React and our web mapping library GL JS. The dashboard polls the API gateway and updates 150 assets every second. While it will require some performance adjustments at scale, you can track and visualize thousands of assets in real-time.
The real value of tracking isn’t just knowing live-location, it’s also the insights you collect and analyze over time — the ability to pair historical data with a real-time view of your operations. For a small taste, check out Kepler, the suite of geospatial analysis tools Uber built with our Custom Layers API to analyze billions of live location updates. We’re making the Custom Layers API available to everyone this summer so developers can extend our mapping libraries with sophisticated WebGL visualizations, going beyond circles, lines, and polygons.
Learn more about our Transportation & Logistics industry solutions for asset tracking and other features like distance and ETAs, live traffic, map-matching, and geo-fencing. Reach out to our team with any questions.
The four things you need to track anything in real time was originally published in Points of interest on Medium, where people are continuing the conversation by highlighting and responding to this story.