30 / Rotman Management Spring 2017
that have been a key part of offline retail and warehouse man-
agement for decades. Machine-learning technologies made
these predictions better, and over the past two decades, much
of the rest of the fulfillment process has been automated.
For example, research determined that fulfillment centre
workers were spending over half of their time walking around
the warehouse to locate items that had been ordered, pick them
off the shelf and put them in a tote. As a result, several com-
panies developed an automated process for ‘bringing shelves
to workers’. Amazon acquired the leading company in this
market, Kiva, in 2012 for $775 million, and eventually stopped
servicing other Kiva customers. Other providers subsequently
emerged to fill the demand for the growing market of in-house
fulfillment centres and ‘3PLs’ (third-party logistics firms).
Despite significant automation, fulfillment centres still em-
ploy many humans. Perhaps surprisingly, the reason is because
of the physical challenge of grasping. Although grasping objects
is easy for humans — infants develop the skill during the latter
half of their first year — this task has so far eluded automation.
It turns out, the core challenge is not in creating dextrous fin-
gers, but in identifying the right angle to use in grasping a par-
ticular object. As a result, Amazon alone still employs 40,000
human pickers full time, and tens of thousands more part time
during the busy holiday season, handling approximately 120
picks per hour.
It is not that companies that do high-volume fulfillment
would not like to automate picking. In fact, for the past two
years, Amazon incentivized the best robotics teams in the world
to work on the long-studied problem of grasping by hosting the
Amazon Picking Challenge. Even though top teams from insti-
tutions such as MIT worked on this problem, as of this writing,
the problem has not yet been solved in a manner that is satisfac-
tory for industrial use. It may seem hard to believe that robots
are perfectly capable of assembling a car or flying a plane, but
are not able to pick items off a shelf and place them in a box.
However, perhaps most surprising is to learn that prediction lies
at the root of the physical task of grasping.
Robots can assemble an automobile because the compo-
nents are highly standardized and the process highly routin-
ized. However, there is an almost infinite variety of shapes,
sizes, weights, and firmness of items in an Amazon warehouse.
Therefore, it is impossible to program robots to pick and place
each and every type of item in whatever orientation they hap-
pen to be positioned in on the shelf. Instead, they must be able
to ‘see’ the object (analyze the image) and predict what ap-
proach to grasping the object will work (arm approach, finger
positioning, grip pressure, etc.) so as to hold it and not drop or
crush it.
In other contexts, however, more abundant prediction
could actually lead to an increased value for human-led tasks,
where the machine provides predictions but the human exercis-
es judgment and undertakes the action component of the task.The Managerial Challenge
As AI technology improves, prediction performed by machines
will increasingly dominate prediction performed by humans.
Managing for such a future requires understanding three inter-
related insights.- PREDICTION IS NOT THE SAME AS AUTOMATION. Prediction is an in-
put to automation, but successful automation requires a variety
of other tasks. The anatomy of a task involves data, prediction,
judgment and action in order to attain an outcome. Machine
learning is only one component of this (prediction). For exam-
ple, medical care tasks include imaging (data), diagnostics (pre-
diction), treatment choice (judgment), bedside manner (judge-
ment/action) and physical intervention (action), among others.
Prediction is the stage of automation in which the technology is
currently improving particularly rapidly, although advances in
sensor technology (data) and robotics (action) are also advanc-
ing quickly. - THE MOST VALUABLE WORKFORCE SKILLS INVOLVE JUDGMENT. For
many activities, prediction was the bottleneck to automation.
This meant a role for human workers in aiding a variety of
prediction tasks, including pick-and-place and driving. Going
Over time, feedback can transform some aspects
of judgment into a prediction problem.