MLS-C01 Question 130
Single answerA retail company wants to optimize its supply chain to reduce delivery times and minimize costs. Their operations team has suggested using machine learning to predict the most efficient delivery routes. However, the team already has access to pre-defined static delivery routes based on historical data and geographic information, which they claim are accurate. When should the team consider using machine learning for this problem?
- A
When the delivery routes are highly dynamic and influenced by multiple unpredictable factors such as traffic, weather, or real-time demand.
- B
When the static delivery routes have been consistently accurate over time and no significant changes are expected in the near future.
- C
When the operations team has sufficient historical data but lacks the computational power to process it using traditional methods.
- D
When the team wants to experiment with machine learning even though the static routes are already meeting business objectives.
Show answer and explanation
Correct answer: A
Explanation
Machine learning is a powerful tool but should only be used when it provides a clear advantage over existing solutions. In this case, the team should consider machine learning if the delivery routes are highly dynamic and influenced by unpredictable factors that static rules cannot handle effectively. If the current static routes are already accurate and effective, there is no need to introduce the complexity of machine learning.
- A. Correct.
This is correct because machine learning is most useful in scenarios where the problem involves complex, dynamic, and unpredictable factors that are difficult to model using static rules or traditional methods.
- B. Incorrect.
This is incorrect because if the static routes are consistently accurate and effective, there is little to no value in implementing machine learning, as it would add unnecessary complexity.
- C. Incorrect.
This is incorrect because computational power is not a direct reason to use machine learning. Instead, the decision to use ML should depend on the nature of the problem and whether ML can provide superior results.
- D. Incorrect.
This is incorrect because implementing ML purely for experimentation without a clear need or benefit is not a justified reason, especially when the current solution is meeting business goals.