At the DataEDGE conference today, Quentin Hardy interviewed Michael Chui, senior fellow of the McKinsey Global Institute, about the Internet of Things (IoT). Representing a management consulting firm, Chui emphasized the business potential of IoT. He maintains that this sector hold great promise.
For many, "Internet of Things" brings to mind consumer products like the Nest thermostat and Fitbit. Chui reported that two thirds of the Internet of Things economy is in business-to-business (B2B) applications, rather than business-to-consumer (B2C) applications. This changes the picture from one of wearable devices to sensor arrays monitoring the oil processing and manufacturing supply chains.
There is a firmness to this picture that brings IoT much more in line with traditional operations research than many admit. Monitoring and tuning supply chains is an ancient business practice. If we see B2B IoT solutions as an extension of this practice of making supply chains more reliable and efficient, then that can put recent trends in perspective.
Chui was optimistic about the value of data generated by both B2B and B2C Internet of Things products and technologies, especially when these systems are linked. The discussants brought up the idea of A/B testing the design of physical objects like consumer products using Internet of Things technologies. In this picture, monitoring of consumers could potentially feed back into the management of the supply chain, not unlike how the monitoring of purchases from retail stores like CVS and WalMart allows these chains to improve their processes for stocking their stores.
What does any of this have to do with data innovation in the social sciences?
The availability of data from social (as well as mechanical) behavior that Internet of Things promises comes with ample opportunities for the data literate social scientist. But while analysis of this data will benefit from computational tools and statistical methods like we teach at the D-Lab, commercial applications of IoT data will be agnostic if not atheistic about any particular disciplinary lens. The Internet of Things will increase demand for robust prediction and control in the service of pragmatic ends.
This kind of pragmatic application of scientific methods to data that is sometimes socially generated has come to be called "data science". It remains to be seen whether the disciplinary formation of academically trained social scientists is relevant to the challenges faced by data scientists in a commercial setting. One possibility is that disciplinary lenses will provide the kind of theoretical insight that enables efficient preprocessing and intelligent modeling of data collected from the ecologically instrumented field. We can expect these pragmatic concerns to put the scientific power of these disciplinary lenses to the test.
At the D-Lab, we teach today's research methods--which come from across the many social science disciplines--in an undisciplinary way. This gives our students a versatility which can give them a wider range of ways to apply their scientific skills. Some of our students will find themselves prepared to research the Internet of Things.
