The wanton stings and motion of the sense

Shakespeare may not have had the Internet of Things (IoT) in mind when he wrote Measure for Measure but the IoT is all about motion and sensors. And measures for that matter.

When you think about the IoT you tend (or I did, at any rate) to think about familiar technologies from familiar vendors. To store data you think about relational databases with capabilities like time series and geospatial capabilities and/or you think about NoSQL databases like Hadoop. On the other side of the coin you think about the mainstream business intelligent vendors running queries against data held in these platforms. However, this isn’t the only option.

The other side of the IoT coin are the sensors that are collecting the data and perhaps it should come as no surprise that vendors with a history in this space, who have collected and processed sensor data for years, should now be applying analytics to this data. One such is Savi.

Savi Technology has been in the sensor business one way or another for 25 years and it makes tags, readers and RFID technology. It also provides a number of applications such as supply chain management and asset tracking. The application I am interested in, however, is Savi Insight, which provides a predictive and prescriptive (operational) solution for supply chain management and logistics. It provides things like in-transit asset visibility, fleet optimisation and risk management.

Now, I don’t usually get into vertical solutions but Savi Insight (which is delivered as an end-to-end SaaS application) has potentially broader uses. It collects information from sensors (its own or others’; including GPS data, weather sensors, social sensors [Twitter and so forth]), it has time series support, has geolocation capabilities, and provides real-time monitoring based on an event processing engine (Apache Storm).  It supports a variety of graphical visualisation tools as well as the ability to overlay information onto maps. It also offers different varieties of dashboard for, for example, business users as opposed to data scientists. The product can also be integrated with application environments such as SAP.

Put all of that together and you have the makings of a general-purpose sensor-based platform for predictive and operational analytics. Now, Savi may not choose to address the general-purpose market but it might well make sense for the company to offer its platform to partners who can build their own applications. This is, in effect, is what Savi itself has done: building pre-built scenarios and patterns that support the supply chain and logistics markets that it targets.

If you are at all interested in this space, either because you are concerned about supply management and logistics, or because you see yourselves as a potential OEM partner of Savi, then I recommend it. Certainly it is worth having a demonstration: most of these that I see are pretty much ho-hum but Savi Insight is impressive.