← All workshopsRead Which Designs Are Actually Acceptable?

Compare Designs Under Constraints

Keep only acceptable inspection models

Tighten the permitted miss count

Botie has measured candidate latency and misses on comparable fixtures. A hard failure cannot be compensated by excellence on the other measurement.

BOTIE

With at most two misses and a 100 ms deadline, no supplied candidate passes both. What follows?
The inputs we are comparing

Latency statistic by candidate (ms)

Same in both cases

  1. 060
  2. 190
  3. 2120
  4. 380

Misses on the same faulty-part set

Same in both cases

  1. 05
  2. 13
  3. 21
  4. 34

Maximum latency (ms)

Same in both cases

100

Maximum miss count

Two feasible candidates

4

Tighter miss limit

2
Result for two feasible candidates
  1. 00
  2. 11
  3. 20
  4. 31
Choose an explanation to test
The fastest candidate misses too many faults

This check uses the pieces from “Require each candidate to pass both limits”.

BOTIE

Botie evaluates four candidate systems. A candidate is feasible only if latency is at most 100 milliseconds and missed faults are at most 2 on this evaluation set. Return a flag for each candidate in the supplied order.
Latency statistic by candidate (ms)
  1. 040
  2. 1100
  3. 2130
  4. 390
Misses on the same faulty-part set
  1. 06
  2. 12
  3. 20
  4. 33
Maximum latency (ms)
100
Maximum miss count
2

Use a number, text, or an array matching the requested result. For a matrix, put each row in brackets. Saving your prediction runs your current construction on these inputs and checks the earlier examples too.

Build a construction in the workspace above first. This case will test those pieces.

What additional evidence could justify changing the candidate set or revisiting a requirement?

Filter by hard requirements before trading off acceptable designs, and preserve uncertainty behind the measurements.

These sample measurements do not establish future guarantees. Compare uncertainty, operating conditions, and any omitted requirements.

Work through the Python challenges →
Continue the courseWhich Assumption Could Change Our Decision? →
← Should the Prediction Happen Here or on a Server?