At the end of my 2nd year of engineering, during the summer break, I signed up for a 3 month internship at the NPCIL — Kakrapar Atomic Power Station. Anyone who has ever worked at a Power Plant will tell you that after working over there, almost everything else is a let-down. The sheer power, elegance, scale and complexity of things there make you feel petite and puts the notion of enormity into perspective.
My reporting authority over there was the Dy. Plant Director, Mr. Rajendran. He was the kind of man who would know things; the kind of man you could depend upon; the kind of man, who, if in-charge, would only give you peace of mind.; You know, just plain sorted.
A week post my joining, the plant was giving an output of about 216 MWe against the 220 MWe it was initially set up for in 1995. My job at the plant was to understand the thermodynamic cycle it worked on and take 6 measurements of about 19 equipments each day. Not of much consequence, to be honest. As all things built in 1995, the readers were analogue vs. the beautiful and accurate digital readers that we have today. This would mean that it would be harder to figure out the flaw in the whole cycle.
There were a lot of meetings held to figure out the issue. Sometimes, twice in a day but 5 days had passed and there was no major break-through. The plant director ordered an audit into all the changes that were made in equipments the past year. In meetings post, it was discovered that 3 months ago, they had undergone an exercise of changing all the pumps at the plant. These pumps, the size of a heavy duty truck, were used to take hot steam up the draft tunnels and convert it back to water. If the pumps were causing heat-loss, that could be a probable cause for lower output.
I remember one conversation during a meeting distinctly:
Plant Director asked impatiently “How do we prioritise the investigation? If NPCIL HQ finds out and it hasn’t been fixed by then, they are going to literally kill us!”
“Well, we start with all the measurement equi.. “ Mr. Rajendran was interrupted by the plant director “What are we doing about the pumps?” He asked.
“But first, we need to fig..” The plant director again interrupted while standing up “I need a report on all the 17 pumps by tomorrow morning. We’ve got to fix this ASAP.” And then he left.
Me, a couple other interns and about 21 people who worked on the floor looked up to Mr. Rajendran for next steps.
“It’s been made pretty clear.. we start with pumps then..” Said Mr. Rajendran with evident lack of confidence. He assigned people to the task and ended the meeting.
About 15 mins post the meeting, he said “You know? Post hoc, ergo propter hoc?” All of us were confused. We weren’t even sure if these words meant something or even existed, for that matter!
Looking at our faces, he added “Post hoc, ergo Procter hoc is Latin for ‘after this, therefore because of this’”. “It’s a logical fallacy. Just because event A follows event B, we think event B caused event A. But that’s not how things happen. There’s no evidence for correlation between change in pumps (event B) with low output of the plant (event A) but since the plant’s low output happened after the change in pumps, we think that’s the most probable cause. It must be investigated but It deserves no priority or special attention over every other probable cause.” That was my first tryst with a logical fallacy. Up until that, I was sold on investigating the pumps as a probable cause.
I was so sold that I could literally observe my opinion changing in my head!
After that, I looked up a host of logical fallacies online and been catching up on videos about them as well. This has shaped a whole lot of decision making since. I wanted to share this story with you in hopes of you also looking for such logical fallacies and reading about it. Once you get to know them, you also will understand how to not let them govern your behavior. It’s a powerful piece of knowledge to have. You can look up more fallacies here.
Also, after 3 days of measurements, it was discovered that the equipment that showed the output itself was flawed. The plant was generating 220MWe the whole time and the reason why NPCIL HQ was not aware of the problem was because there was none.
Until next time.