Discussion – 

0

Discussion – 

0

Data Deep Dive – What does your battery tell about you?

Enterprise workers, and the IT teams that support them, know that enterprise mobility’s success hangs on batteries in each device being healthy and able to last an entire shift of work – and Elemez Battery Essentials provides information exactly for that purpose. With Battery Essentials you can see in one view if there are any batteries that do not hold charge any more and need replacing or if there are devices that show problematic high discharge rates that need investigating. With that information operations managers can pro-actively manage their batteries and prevent failures and downtimes before they happen. (Click here for more information) 

But battery data provides so much more information about a device and its usage. Researchers from Stanford University have recently shown that by monitoring small changes in the battery discharge of a mobile device they could track its location. When a device has a poor network connection, for example because it is farther away from the next mobile signal mast, it will use more power to keep up its connection. Using machine learning techniques the Stanford researchers could determine which route a device must have taken given the changes in battery discharge over time.  

Don’t worry, the data Elemez collects on battery usage is not on a granularity that would allow us to replicate the Stanford results (if you want Elemez to track location you will have to explicitly switch on GPS location tracking via the configuration). But short of such an application from the world of spies and secret agents, the battery still provides us with plenty of valuable information.

At B2M we process over 150 million individual battery status updates per day  across over 1.5 million devices.  A few of the insights on end-user behaviors we can gain from this data:

Charging behaviour: The best battery will not get you through the day, if it has not been charged. Elemez can tell you exactly when devices have been put on charger and when they have been taken off. Are devices put on charger over night? Do workers have to recharge during shift times to get through the day? If devices should be put in a dock when in a van/truck/lorry – are they?

Faulty chargers: A loose connection with the charger can mean a device flickers between on- charge and off-charge. Connecting to a charger often wakes up devices and might even light up the screen. A charger with a loose connection might therefore not only be less efficient in charging a mobile device, it could actually end up draining the battery even faster. Elemez can help you detect such problems with chargers, so they can be repaired or replaced.

Work hours: Often we care about monitoring the use and health of devices during work hours (e.g. we might care about the mobile signal strength when a device is out and about, but not when returned to a depot at the end of a shift). Most operations managers probably know when devices are supposed to be used. But are they actually being used during those hours? At B2M we work to give you relevant insights on all of this across large device deployments spanning different time zones, with devices constantly moving between locations without you having to dig into complicated configurations. Battery data (together with other metrics) gives us valuable context on the use of devices.

Problem applications: Companies need to be agile and move fast in today’s world. Continuous deployment means new versions of applications can go into production at any time and change often. Increasing battery discharge can be an early indicator of a new application version misbehaving and causing problems on the devices. If a new version unexpectedly drains the batteries of devices faster then the old application, it uses more resources on the devices, which could cause devices to slow down and applications to crash. Avoiding errors in all cases is not achievable, but if we detect them early we can rollback and get them fixed before frontline workers call the support desk and our operations come to an halt.

Do we have all the answers in our data? Usually, but as our customers deploy our solution, and learn to depend on the real-time insights we provide, they constantly ask “can we also learn / see……”. As a company, we provide the tools and data expertise to work with our partners to understand their questions. As our CEO has said previously “we are not a supplier, we are a partner.” 

As data scientist at B2M what motivates me most is when I hear from our customers about the actual problems they are trying to solve, and work with them to figure out how analytics can help them. In this blog post I tried to give you some ideas about the power of analytics related to batteries.  I would hope to hear from you what challenges we can tackle together using our real-time analytics!

 

Note:  For additional information on the Stanford study, click here

Tags:

Anton Flugge

0 Comments

Submit a Comment

Your email address will not be published. Required fields are marked *

You May Also Like