Quandaries and Queries


In the January 11,1988,issue of the Oil&Gas Journal, R.A.Baker describes how the Bayesian approach can be used to revise probabilities that a prospect field will produce oil. In one case he describes, geological assessment indicates a 25% chance the field will produce oil. Further,there is an 80% chance that a particular well will strike oil given that oil is present on the prospect field.

  1. Suppose that one well is drilled on the field and it comes up dry. What is the probability the prospect field will produce oil?
  2. If two wells come up dry, what is the probability the field will produce oil?
  3. The oil company would like to keep looking as long as the chances of finding oil are greater than 1%. How many dry wells must be drilled before the field will be abandoned
  4. If the first well produces oil,what is the chance the field will produce oil?


Hi Wei,

Let event B={a field will produce oil} and event Ai = { i-th drilled well will produce oil}.
It is given that P(B)=0.25 and P(A|B)=0.8. Hence P(not Ai|B) = 1-0.8=0.2 for any i=1,2,... and P(not B) =1-P(B) =0.75. Moreover, it is obvious that P(not Ai| not B)=1 for any i=1,2,...

We made two assumptions to solve this problem. The first is that if oil is present in the field then the field is capable of producing oil. The second is that the probability that a particular well produces oil is independent of whether any previous well has produced oil, that is the events Ai are independent.


We need to calculate
P(B| not A1)= Bayes formula = P( not A1|B)*P(B) / P(not A1).
To calculate P(not A) we use the formula of total probabilities:

P(not A1)
= P(not A1|B)*P(B) + P(not A1|not B)*P(not B)
= (0.2)*(0.25) + (1)(0.75)
= 0.8.

Substituting all numbers we obtain:
P(B| not A1)= (0.2)*(0.25)/0.8 = 0.0625=6.25 %


We need to calculate  
P(B| not A1 and not A2)= Bayes formula = P( not A1 and not A2| B)*P(B) / P(not A1 and not A2).
To calculate P( not A1 and not A2| B) we use independence:
P( not A1 and not A2| B)= P( not A1| B)*P(  not A2| B) =(0.2)*(0.2)=0.04.
To calculate P(not A1 and not A2) we use the formula of total probabilities:

P(not A1 and not A2)
=  P(not A1 and not A2|B)*P(B) +  P(not A1 and not A2| not B)*P(not B)
= (0.04)*(0.25) +(1)*(0.75)=0.76.

Substituting all numbers we obtain: 
 P(B| not A1 and not A2)=  (0.04)*(0.25) /(0.76)=0.0132 = 1.32%


Now consider three wells and calculate  P(B| not A1 and not A2 and not A3)
If this is less than 1% then three well will suffice. If not try four wells.


Certainly, it will be 1 (common sense). Is that what you get when you apply Bayes Formula?

Andrei and Penny


Go to Math Central