AIF-C01 Question 27
Select 3A retail company wants to optimize its inventory management system to avoid stockouts and overstock situations. The team is considering using a machine learning (ML) model to predict future inventory needs. However, after analyzing the situation, they realize the cost of implementing the ML solution outweighs the potential benefits. In which of the following scenarios would using an AI/ML solution NOT be appropriate?
- A
The company has a small, predictable product catalog with stable demand patterns.
- B
The data required for training the ML model is incomplete and expensive to collect.
- C
The company needs a deterministic, rule-based system to meet compliance requirements.
- D
The company is experiencing highly volatile demand and wants to improve prediction accuracy.
- E
The company has a large dataset and wants to discover hidden patterns to optimize inventory.
Show answer and explanation
Correct answers: A, B, C
Explanation
AI/ML solutions are not always appropriate, especially when simpler or more cost-effective approaches can achieve the desired outcomes. Scenarios where stable demand patterns exist, data is expensive to collect, or deterministic results are required are better addressed with non-ML solutions. In contrast, AI/ML is better suited for discovering hidden patterns in large datasets or making predictions in highly volatile environments.
- A. Correct.
Using an AI/ML solution is not appropriate when the product catalog is small and demand patterns are stable because a simple rule-based or statistical approach can provide sufficient accuracy at a lower cost.
- B. Correct.
If the data required for training the ML model is incomplete and expensive to collect, the cost of data preparation and collection can outweigh the benefits of implementing the solution.
- C. Correct.
A deterministic, rule-based system is more suitable for compliance requirements, as AI/ML models are probabilistic and may not guarantee specific outcomes.
- D. Incorrect.
Highly volatile demand patterns might benefit from an AI/ML solution if the goal is to improve prediction accuracy, making this scenario appropriate for an ML approach.
- E. Incorrect.
A large dataset with hidden patterns is a strong use case for an AI/ML solution, as ML models excel at uncovering these patterns to optimize processes.