# Estimation from image

Bain Case: Old Winery estimation picture
Neue Antwort am 18. Dez. 2019
3 Antworten
11,7 Tsd. Aufrufe

Hello,

I believe estimating values (vines per area) from the image only to be a nigh impossible task. Everyone I have given this case to struggles with this. Any advice on how to proceed?

Regards

## Übersicht der Antworten

• Datum aufsteigend
• Datum absteigend
• Beste Antwort

I think this particular question is not designed to reach an accurate result. There are simply too many assumptions involved, and their errors add up. Instead, the question tests whether the candidate can logically develop a reasonable approach to this problem, and should be treated as such.

War diese Antwort hilfreich?

Hello,

You are right, it's quite impossible to make a reliable estimation.
That being said, you have to enter the logic that your interviewer is not interested in the answer, sometimes he doesn't even know the answer. This is a perfect example of when you interviewer wants just to test your problem solving skills and see how you work under pressure.

In these cases, you have to:

1. Deline a strategy and expose it to your interviewer, making clear each step and being really careful to his potential inputs
2. Estimate each number doing an educate guess, everytime involving the interviewer in your reasoning
3. Do the calculations

Point 2 is crucial and it doesn't matter what is the number that you propose: it has only to be justified in a rational way.

Hope it helps,
Luca

War diese Antwort hilfreich?

1. you can quickly estimate the row spacing by using the height of an average human (pictured) and then
2. eyeball the size of a leaf (in case you've never seen one in real life) and from there estimate the size of the grape bunch in relation to that
3. assume that grapes have a relatively similar density to water, and with the size of the bunch estimated you can then ballpark its weight

again, there are a few different ways to do this but it should be just done quickly so you can focus on the core problem.

Regards,

War diese Antwort hilfreich?
Wie wahrscheinlich ist es, dass du uns einem Freund oder Kommilitonen empfiehlst?
0
1
2
3
4
5
6
7
8
9
10
0 = Nicht wahrscheinlich
10 = Sehr wahrscheinlich