Lately, I’ve seen a surge of new offers for real applications that use sensor data to deliver a valued service. This has me thinking about how data warehouses are primarily oriented towards the transactions of people – phone calls, grocery purchases, web browsing… all of this relates to people and their daily actions.
Going on around us, almost invisibly, is another world, the world of “things.” Things like cars, washing machines, heat pumps… Today there are very few “things” that do not have some type of sensor that generates data. Sensors can measure temperate, motion, location, light, acceleration, current, radiation, flows, biometrics e.g., heart rate or pulse, oxygen levels, RFID, etc. Today many of our “things” are in fact internet enabled or can be.
Hence the term “Internet of Things”.
The Internet has enabled communications between people and Things very well. What’s missing? Well, how do we address the data of Things and analyze it in a valuable, meaningful way? How do we correlate the data output from Things and relate that to customers/people? It is possible and I will share two examples that I think show the potential of using a data warehouse to …
Lately, I’ve seen a surge of new offers for real applications that use sensor data to deliver a valued service. This has me thinking about how data warehouses are primarily oriented towards the transactions of people – phone calls, grocery purchases, web browsing… all of this relates to people and their daily actions.
Going on around us, almost invisibly, is another world, the world of “things.” Things like cars, washing machines, heat pumps… Today there are very few “things” that do not have some type of sensor that generates data. Sensors can measure temperate, motion, location, light, acceleration, current, radiation, flows, biometrics e.g., heart rate or pulse, oxygen levels, RFID, etc. Today many of our “things” are in fact internet enabled or can be.
Hence the term “Internet of Things”.
The Internet has enabled communications between people and Things very well. What’s missing? Well, how do we address the data of Things and analyze it in a valuable, meaningful way? How do we correlate the data output from Things and relate that to customers/people? It is possible and I will share two examples that I think show the potential of using a data warehouse to analyse and understand Things!
First off my current favorite – Fedex SensorAware.
This is a device that measures location (via GPS), acceleration, temperature and light and can transmit this data over a mobile data network from “inside” your package. This enables tracking of your package and monitoring of what can be important environmental variables. You don’t want frozen food to thaw, or a disk drive to be dropped
And the London Oyster card.
Likely the world’s largest and most successful RFID card implementation with over 10 million cards in circulation. Used for something like 5 Million trips per day in the greater London area. It has largely replaced cash ticket purchases.
This card is a “prepaid” train and bus ticket. It can be topped up at kiosks or online. For the big brother worriers, today you don’t need to register your name and can use it anonymously. However. if you do register you can view all your trips, top-up on-line and freeze your card if you loose it and get a new one with the remaining balance.
The data from Oyster can be used for transport and route planning.
There are many more examples out there in use today. Data from Things is an opportunity to differentiate data warehousing solutions and include new data sources to deliver better business outcomes for all.
The sensor data can be very usefully analysed by statistical analysis tools to provide previously unknown insights, correlations and relationships.
So I suggest you do have a good think about the Things out there, which can be a valuable data source, as well as the customer transactions when building a data warehouse. You may well be surprised!
