Business Analytics, Big Data Analytics interview case study

Contents

Introductionroad-confussion-2695280

The case study is the most important round for any analytics engagement. But nevertheless, a lot of people get nervous at the mention of undergoing a case interview. There are many reasons for this, but the most popular are:

  • You need to think about your feet in a situation where there is already enough pressure
  • Limited resources available to prepare for analytical case studies. Even with the amount of content available on the web, not many analytical case studies are freely available.

From an interviewer's perspective, is judging the candidate based on structured thinking, problem solving and comfort level with numbers using these case studies. This article will take you through a case study. The answer to each question leads you to delve into the same problem.

Be sure to check the ‘Interviews in Ace Data Science‘ course. We have poured all of our combined experience of over a decade and hundreds of interviews into this comprehensive and definitive course.. It's a guide you don't want to miss!!

Bottom:

I moved to Bangalore ago 10 months. Bangalore is a large city with several roads labeled one-way. You make a mistake and are late more than 20 minutes. Every day I compare the time spent on different routes and choose the best among all possible combinations. This article takes you through an interesting riddle that took me considerable time to solve..

Process to solve:

I have structured it in a very similar way to an analytical interview. Background information will be provided at the beginning of the interview, which will be followed by questions. Once you've brainstormed / solved a question, you will be presented with additional information that will advance the case further.

If you want to go through this case in true spirit, just ask one of your friends to take the questions and information (provided in the next section) and present them to you at the right time. After all the questions, I have provided the answers I expected. You can compare your answers with mine.

Note that there is no right or wrong answer in many situations and a case evolves the way the interviewer wishes. If you have an answer / different approach, feel free to post in the comments and I'd love to discuss them.

Problem Statement :

Bottom : There are two alternative routes that I take to get to the main road from my house.. The average speed on each of the roads is about 30 km / h. Let's call the two roads as Highway A and Highway B. The total distance needed to travel on Highway A and Highway B is 1 km y 1,3 km respectively to reach the same point of the main road. Note that, before the two roads divide, i see a sign (for instance, WITH) which is common to both roads and, Thus, not included in this calculation. See figure for clarification.

roads1-8397701

P1: What are the possible factors that you should consider in calculating the total time spent on each path??

P2: Which way should you take to get to the main road in order to minimize the time required? And what is the difference in the total time taken by the two alternative routes?

Additional Information (will be provided after the question 2): Recently, one of the crosses (Let's say, X) on highway A it got too crowded and a traffic light was installed on it. The traffic signal was configured to 80 seconds in red and 20 seconds green. Let us denote the signal seconds as R1 R2 R3… G1 G2 G3. Here, R1 denotes 1 second after the signal turns red.

P3: Does it still make sense to take Highway A or switch to Highway B as long as the average speed on Highway A remains the same?, except at the stop at the sign?

Additional Information (will be provided after the question 3): If I get to the signal on R1, I'll be in the front rows to be released once the sign turns green. While, if I reach the signal in R80, I may have to wait a while even after the signal turns green because the vehicles in the first rows will block me for a few seconds before starting. Let's make some realistic guesses about the wait time after the signal turns green.

R1 – R 10: 0 seconds, R11-R20: 3 seconds, R21 – R60: 10 seconds, R61 – R80: 15 seconds, G1-G15: 5 seconds, G15-G20: 0 seconds

P4: Does it still make sense to take Highway A or switch to Highway B as long as the average speed on Highway A remains the same?, except at the stop at the sign?

P5: Can you think of a reason why Highway A may be a better option to get to Junction X in minimal time??

Additional Information (will be provided after the question 5): The Z signal (before the two roads divide) has exactly the same cycle as the signal at point X, namely, 90 seconds in red and 20 seconds green. The average speed of any vehicle varies on Highway A from 25 km / h (heavy traffic) a 30 km / h (light traffic). The X signal is shifted from the Z signal by 25 seconds. Therefore, when it turns green in Z, is R55 in signal X.

P6: Does it still make sense to take Highway A or switch to Highway B as long as the average speed on Highway A remains the same?, except at the stop at the sign?

Solution :

Bottom : There are two alternative routes that I take to get to the main road from my house.. The average speed on each of the roads is about 30 km / h. Let's call the two roads as Highway A and Highway B. The total distance needed to travel on Highway A and Highway B is 1 km y 1,3 km respectively to reach the same point of the main road. Note that, before the two roads divide, i see a sign (for instance, WITH) which is common to both roads and, Thus, not included in this calculation.

Question : Which way should you take to get to the main road to minimize the time needed? And what is the difference in the total time taken by the two alternative routes?

Solution:

Time spent on highway A = 1/30 * 60 min = 2 minutes

Time spent on highway B = 1,3 / 30 * 60 min = 2,6 minutes = 2 min 36 seg

Therefore, the clear choice is highway A. Highway B would have taken 36 seconds longer than highway A.

The interviewer tests your comfort with numbers and your confidence with the answer in this step.


Bottom : Recently, one of the crosses (Let's say, X) on highway A it got too crowded and a traffic light was installed on it. The traffic signal was configured to 80 seconds in red and 20 seconds green. Let us denote the signal seconds as R1 R2 R3… G1 G2 G3. Here, R1 denotes 1 second after the signal turns red.

Question : Does it still make sense to take Highway A or switch to Highway B as long as the average speed on Highway A remains the same?, except at the stop at the sign?

Solution: Suppose I hit the signal at a random time. Therefore, the probability of reaching the signal at R1 R2 R3… or G1 G2 G3 are all equal. Therefore, the expected time in the signal is:

E (downtime) = (1 + 2 + 3 + 4 + …… .80) / (80 + 20) = (80 * 81) / (100 * 2) = 32.4 seconds.

We still see 32,4 seconds <36 seconds. Therefore, it still made sense to take highway A.

The interviewer tests your knowledge of statistics (Calculation of expected value), the approach to the problem and the interpretation of the final results in this step.


Bottom : So far, the solution will look good in the books. Spice up the problem with basic realities. If I get to the signal on R1, I'll be in the front rows to be released once the sign turns green. While, if I reach the signal in R80, I may have to wait a while even after the signal turns green because the vehicles in the first rows will block me for a few seconds before starting. Let's make some realistic guesses about the wait time after the signal turns green.

R1 – R 10: 0 seconds, R11-R20: 3 seconds, R21 – R60: 10 seconds, R61 – R80: 15 seconds, G1-G15: 5 seconds, G15-G20: 0 seconds

Question : Does it still make sense to take Highway A or switch to Highway B as long as the average speed on Highway A remains the same?, except at the stop at the sign?

Solution:.

E (downtime) = {(1 + 2 + 3 + 4 + …… .80) + 3 * 10 + 10 * 40 + 15 * 20 + 5 * 15} / (80 + 20) = 40.15 seconds.

This time the game changes and how 40,15 seg> 36 seg, I will prefer road B over road A.

Interviewer tests how quickly you change some of the assumptions to minimize aggregate calculations.


Bottom : Even after doing such a logical calculation, noted that in 30 different events, traveled more than 25 seconds faster on highway A compared to highway B every time. I did not change my average speed on any of the roads. It might have been acceptable in case it found x number of events where A wins and 30 – x where B wins. But winning every time was suspect. I've been fighting for the last 10 days to find valid cause. It caught my eye today and the following is what I discovered:

The Z signal (before the two roads parted), which I initially thought had nothing to do with the calculation, it was actually the game changer. This is how it played a role. This signal had exactly the same cycle as the signal at point X, namely, 90 seconds in red and 20 seconds green. As long as the two lights have the same cycle, the incidence on signal X is no longer random.

Question : Does it still make sense to take Highway A or switch to Highway B as long as the average speed on Highway A remains the same, except at the stop at the sign?

Solution:

Let's say my average speed varies on Highway A from 25 km / h a 30 km / h. The X signal is shifted from the Z signal by 25 seconds. Therefore, when it turns green in Z, is R55 in signal X.

Case 1: (Light traffic) Time needed to travel the road A = 2 minutes = 120 seconds

Reading in X when I reach the signal = R55 + 120 = R75.

Case 2: (Heavy traffic) Time needed to travel the road A = 2 minutes 24 seconds = 144 seconds

Reading in X when I reach the signal = R55 + 144 = G19

Therefore, the probability of R1-R74 is zero. And the revised equation for the expected time is:

E (downtime) = (5 + 4+ 3+ 2+ 1 + 15 * 5 + 5 * 15) / 25 = 6.6 seg

Therefore, What 6.6 seconds <36 seconds, road A always wins on road B.

Therefore, the random event assumption is not always true. Try to find out all the possible factors that could possibly influence the occurrence of an event before making a random event assumption.

The interviewer tests your innovative thinking, questioning your ability to guess and your ability to interpret the results at this step.


Final notes:

Was the article helpful to you? Share with us any other problem statements you can think of. Let us know your thoughts on this item in the box below..

In one of the next articles, we will share how an interviewer judges an analyst during a case study.

If you like what you have just read and want to continue learning about analytics, subscribe to our emails or like ours page the Facebook.

Subscribe to our Newsletter

We will not send you SPAM mail. We hate it as much as you.