Statistical fluctuations and dependent events | MBA Learnings

Current student Rohan Rajiv is blogging once a week about important lessons he is learning at Kellogg. Read more of his posts here.

In my last MBA learning post, we dove into the idea of managing queues by managing our utilization. In that post, we briefly discussed why queues form. Today, I’d like to dig deeper into that question.

So, why do queues / delays occur? They occur due to a combination of statistical fluctuations and dependent events. In simpler terms, statistical fluctuations can be described as variability. If we go back to the analogy of the ice cream stall, it is unlikely that the queue is exactly five people through the day. There will be times when the queue will be long and then times when it will be short. These differences are statistical fluctuations.

Life in the world of managing queues becomes harder when you combine statistical fluctuations with dependent events. A beautiful example of this combination at play is demonstrated in this 1 min 46 sec video called “The Subway Delay Story” by the MTA, New York’s subway operator.

In this video, the sick passenger requiring assistance and an emergency stop is an example of variability. And, every train behind the held train is a dependent event.

Another simple example of statistical fluctuations and dependent events at play is in connecting flights. We all have probably experienced delays in one flight that results in a chain reaction of delays in other flights. As a result, most airlines experience a deteriorating performance in their arrival times as the day progresses.

So, what can we do about statistical fluctuations and dependent events? Well, books have been written on the subject, So, instead of attempting a comprehensive answer, I’m going to share my simplified set of three ideas you can apply immediately in your life:

1. Be sensitive to the combination of statistical fluctuations and dependent events in your own life. 
One simple application would be to avoid scheduling back-to-back meetings through the day. It is inevitable that one of them will spill. And, this, in turn, will result in a chain reaction through your day.

2. Control for statistical fluctuations / variability by building in safety capacity. 
This is a continuation of the idea of reducing your “utilization” and building in safety capacity. So, if you are scheduling many meetings during the day, leave little gaps in between so you can catch up on delays caused by fluctuations.

3. Look for creative ways to reduce variability. 
Disney offers “fastpass” tickets to various rides that allow customers to skip the queue at these rides at certain times. These times are designed such that groups of customers are incentivized to go to these rides at different periods during the day – thus, reducing the likelihood that everyone shows up at the same peak time. The equivalent in our personal life would be to plan ahead and avoid deliverable “peaks.” If you foresee two projects coming due in the same week, finish one up a week earlier and treat yourself (equivalent to the fastpass) so you avoid the peak rush.

And, finally, look out for statistical fluctuations and dependent events in the coming week. It changes the way you think about queues. And, if that isn’t magical, what is?

Rohan Rajiv just completed his first year in Kellogg’s Full-Time Two-Year Program. Prior to Kellogg he worked at a-connect serving clients on consulting projects across 14 countries in Europe, Asia, Australia and South America. He blogs a learning every day, including his MBA Learnings series, on www.ALearningaDay.com.