Which combination of steps should the company take to meet these requirements with the LEAST operational overhead?
(Choose two.)
Develop the model by using the Amazon Forecast Prophet model.
Develop the model by using the Amazon Forecast holidays featurization and weather index.
Deploy the model by using a canary strategy that uses Amazon SageMaker and AWS Step Functions.
Deploy the model by using an A/B testing strategy that uses Amazon SageMaker Pipelines.
Deploy the model by using an A/B testing strategy that uses Amazon SageMaker and AWS Step Functions.
Explanations:
While using the Amazon Forecast Prophet model can be effective for time series forecasting, it does not inherently consider holidays and weather conditions, which are critical for demand prediction in this scenario.
Developing the model using Amazon Forecast’s holidays featurization and weather index allows the company to incorporate relevant external factors, which are essential for accurate demand forecasting based on holidays and weather conditions.
Deploying the model with a canary strategy using Amazon SageMaker and AWS Step Functions allows for gradual rollout and testing of the model in a live environment. This approach minimizes risk and operational overhead while ensuring adaptability to supply chain needs.
An A/B testing strategy using Amazon SageMaker Pipelines might add unnecessary complexity and operational overhead compared to simpler deployment strategies. Moreover, A/B testing typically compares two models rather than testing a single model over a short duration.
Although deploying with an A/B testing strategy using Amazon SageMaker and AWS Step Functions may provide some benefits, it introduces more complexity and operational overhead than needed for a 2 to 3-day test of a single model, making it less suitable for the given requirement.