“What doesn’t kill you makes you stronger” is the advice we often get at life’s difficult moments. The other bit of mildly needling advice is that “it’s not what happens, it’s how you deal with it that counts.” It turns out that science underpins these truisms, and moreover, there are important lessons for economies, corporations and individuals to be learned.
Economist @TimHarford recently presented a fascinating thesis at JDA FocusConnect 2017. His proposition is that the data supports these notions and they turn out to be economically important. In summary, it is important to fail or least to put yourself in the position where failure is a possibility as often as possible. Why this is important goes back to Charles Darwin; fast adaptors get ahead of their competition. If you fail to put yourself in learning situations, then you stand the chance of swimming in the backwater of evolution’s slow lane. Of course, there will be less frequent visits to life’s casualty department if you can prepare (at least some of the time) in the sand-pit rather than through the more bruising real-life encounters we euphemistically refer to as “learning experiences.”
JDA FocusConnect is a supply chain event, so how might Tim’s assertions be applied? And can there be a way to gain with less pain? For example, is it possible to prepare for what Donald Rumsfeld famously called “known unknowns?” The answer: Yes, it is possible. Simulation means alternative supply chain configurations can be modelled against business outcome and strategic conformance. Not only will you become more of an expert, you will also become more agile. Moreover, when the day comes you that you need to put the learning into play in the real world, you can move at speed and with confidence. Then when the “unknown, unknowns” occur, you will have a process in place and embedded learning that enables you to react quickly.
Information services and sensor data from the Industrial Internet/Internet of Things (IoT) out to the edge of the supply chain exponentially increases the available information. JDA Labs is utilizing machine learning to scale up the analytics. For example, in “supervised machine learning,” the input might be weather and the output might be sales. It’s up to the machine to determine how the weather affected sales. Predictive analytics determine when customers will require a replacement product – before the customer even knows and links it to real world triggers. Therefore, the supply side can be prepared in advance. What this also suggests is that predictive analytics and industrial scale learning not only means more agile adaption, but is anticipating the successful moves in advance.
This brings me back to those real-life encounters with experience. It will be a while before we have machine learning linked to our day-to-day encounters. So, the next time you need to console a colleague, family member or yourself because the real-life predictive analytics algorithm misfired, you can remind them that they are officially in life’s fast lane and adapting faster… it’s backed by science.