We use cookies to ensure you have the best browsing experience on our website. Please read our cookie policy for more information about how we use cookies.
Explanation:
Wouldn't it be great if we had the number of products produced for each day so, we could select the MINIMUM day which would produce our goal count?
But we don't. So, do we go ahead and start calculating the number of products produced each day?
No, because time complexity would suffer.
Instead, we use binary search.
The basic logic is that we set a minimum day (lower bound) and a maximum day (upper bound, which would be the time taken by the slowest machine to produce the goal count). Then, we find the "mid" value between these 2 days and see the number of product produced in the "mid" day. If the value is lesser than the goal count, we search for our day between "mid" to maximum day. If the value is greater than the goal count, we search for our day between minimum to "mid" day.
Example: (because I got stuck a lot because of small implementation details)
machines[] = [4L, 5L, 6L]
goal = 12
----begin execution------------
minimumDays = 0
maximumDays = 6 * 12 = 72
So, our search bound is
0(minimumDays)-----------------------72(maximumDays)
Iteration 4:*
19(minimumDays)-------------23(mid)-----------27(maximumDays)
prodCount = 12 == goal (Yay, we found the day that produces the goal! But wait. Could this be the actual MINIMUM day?)
Iteration 5:
19(minimumDays)-------------21(mid)-----------23(maximumDays)
prodCount = 12 == goal (Huh, so this one is lesser than the day we found in the last iteration)
Minimum Time Required
You are viewing a single comment's thread. Return to all comments →
Explanation: Wouldn't it be great if we had the number of products produced for each day so, we could select the MINIMUM day which would produce our goal count? But we don't. So, do we go ahead and start calculating the number of products produced each day? No, because time complexity would suffer. Instead, we use binary search.
The basic logic is that we set a minimum day (lower bound) and a maximum day (upper bound, which would be the time taken by the slowest machine to produce the goal count). Then, we find the "mid" value between these 2 days and see the number of product produced in the "mid" day. If the value is lesser than the goal count, we search for our day between "mid" to maximum day. If the value is greater than the goal count, we search for our day between minimum to "mid" day.
Example: (because I got stuck a lot because of small implementation details) machines[] = [4L, 5L, 6L] goal = 12
----begin execution------------
minimumDays = 0 maximumDays = 6 * 12 = 72
So, our search bound is 0(minimumDays)-----------------------72(maximumDays)
Iteration 1: 0(minimumDays)---------36(mid)--------------72(maximumDays) prodCount = 22 > goal
Iteration 2: 0(minimumDays)-------------18(mid)-----------36(maximumDays) prodCount = 10 < goal
Iteration 3: 19(minimumDays)-------------27(mid)-----------36(maximumDays) prodCount = 15 > goal
Iteration 5: 19(minimumDays)-------------21(mid)-----------23(maximumDays) prodCount = 12 == goal (Huh, so this one is lesser than the day we found in the last iteration)
Iteration 6: 19(minimumDays)-------------20(mid)-----------21(maximumDays) prodCount = 12 == goal (even lesser!)
Iteration 7: 19(minimumDays, mid)-------------------------20(maximumDays) prodCount = 10 < goal
Iteration 8: * 20(minimumDays, maximumDays) * minimumDays == maximumDays, so loop ends. So we return the value from iteration 6! * * */