Use cases for online model serving

Online model serving is essential when we need to see the impact of recent data on the model as soon as possible instead of waiting for a periodic batch update. In this section, we will discuss the following example cases where online serving is needed:

• Recommending the nearest emergency center during a pandemic
• Predicting the favorite soccer team in a tournament
• Predicting the path of a hurricane or storm
• Predicting the estimated delivery time of delivery trucks

Case 1 – recommending the nearest emergency center during a pandemic

Let’s assume a lot of people are becoming sick due to a pandemic, such as Covid, every day. However, there is a limited number of hospitals. We want to recommend the nearest, most convenient hospital to a patient, keeping the following things in mind:

• The distance between the patient and the hospital should be short
• The number of patients sent to a particular hospital...