Olivier Jouve, Director, Predictive Analytics, IBM
By Olivier Jouve
Our world has never been more interconnected and instrumented. Streams of data are constantly being collected from sensors that monitor everything from the environment, vehicles, buildings and bridges, mobile devices and home appliances. And because of these constant streams of Big Data, opportunities exist to effectively predict when equipment will fail, a storm will hit, traffic will increase or milk will spoil.
Human beings are still the most important and sophisticated data generators and sensors; however, our interpretation of signals is still highly unreliable.
That is why companies across every industry are using predictive analytics to collect, assimilate, and analyze the Big Data around behaviors that we humans reveal on a daily basis. By understanding how people act in different environments (offline vs. online) with what they purchase, along with what they say, how they say it, and when they say it, determines what their next action might be. More specifically, it can determine how these actions are directly affecting everything from the manufacture of equipment to inventory levels; and from fraudulent activity to the spread of infectious diseases.
For example, cars are equipped with sensors to monitor the engine, brakes and safety devices, among others, but more and more car manufacturers are incorporating “people data” into how cars are designed and manufactured.
Driving styles and driving environments vary. There is quite a difference between Chicago and Paris. With usage monitoring systems, car companies can better understand how the stress on a set of brakes can impact longevity as well as supply in the auto part aftermarket, and when maintenance, repair and overhaul might be required.
Social media is also an important element in this process. Think back a few years to when a large automotive manufacturer recalled millions of vehicles to fix a host of issues. By listening to the conversations of its customers in the social sphere, it might have been able to catch these issues early, alter the manufacturing process to fix quality issues, be more proactive in responding to customer complaints, and avoid a major crisis communications issue.
The effects of monitoring social media are also extending further into healthcare and having a direct impact on when and how fast diseases spread. Based on people’s tweets last summer, Twitter accurately predicted – eight days in advance and with 90 percent accuracy – when and where people would get sick from the flu.
Human sensors are also impacting the sports world when it comes to staying healthy.
The Leicester Tigers, nine-time champion of the English rugby union’s Premiership, are monitoring all facets of a player’s activity – on and off the field – in order to reduce players’ injury rates. It’s a human form of predictive maintenance.
After all, losing a key player for an extended period of time not only hurts the team on the field, it can result in reduced ticket sales and spectator attendance if the team does not perform up to expectations.
With monitors built into their uniforms, Leicester collects data on fatigue and game intensity levels that allow coaches and trainers to design a personalized training regime for each player tailored to his physical and psychological states, reducing the risk of injury.
Even though humans can often be impractical, impulsive and evasive, their actions speak volumes.
The next time you visit the doctor, climb into your car or tweet an opinion about your favorite brand, consider all of the data you are generating and how that data is being fine-tuned to improve your life – and the lives of those around you.
Please join me on People for a Smarter Planet Facebook page on Wednesday, Jan. 23 for further discussion on this topic: https://www.facebook.com/events/553247638020232/
