PROFESSIONAL-MACHINE-LEARNING-ENGINEER practice question 16 of 35

Your organization wants to make its internal shuttle service route more efficient. The shuttles currently stop at all pick-up points across the city every 30 minutes between 7 am and 10 am. The development team has already built an application on Google Kubernetes Engine that requires users to confirm their presence and shuttle station one day in advance. What approach should you take?

Answer options

A: 1. Build a tree-based regression model that predicts how many passengers will be picked up at each shuttle station. 2. Dispatch an appropriately sized shuttle and provide the map with the required stops based on the prediction.

B: 1. Build a tree-based classification model that predicts whether the shuttle should pick up passengers at each shuttle station. 2. Dispatch an available shuttle and provide the map with the required stops based on the prediction

C: 1. Define the optimal route as the shortest route that passes by all shuttle stations with confirmed attendance at the given time under capacity constraints. 2 Dispatch an appropriately sized shuttle and indicate the required stops on the map Correct answer

D: 1. Build a reinforcement learning model with tree-based classification models that predict the presence of passengers at shuttle stops as agents and a reward function around a distance-based metric 2. Dispatch an appropriately sized shuttle and provide the map with the required stops based on the simulated outcome.

Correct Answer:

C. 1. Define the optimal route as the shortest route that passes by all shuttle stations with confirmed attendance at the given time under capacity constraints. 2 Dispatch an appropriately sized shuttle and indicate the required stops on the map

Explanation

The problem states that users confirm their presence and shuttle station one day in advance. This means the number of passengers and the required stops are known, deterministic inputs. Therefore, the task is not a prediction problem (which would require a machine learning model) but a classic optimization problem. The goal is to find the most efficient path connecting a known set of points, which is a Vehicle Routing Problem (VRP). Option C correctly frames the problem as finding the shortest route that services all confirmed stops, considering vehicle capacity constraints. This is the most direct and appropriate approach.

Why Incorrect

A: A regression model to predict passenger counts is unnecessary because the application provides the exact number of confirmed passengers for each stop. B: A classification model to predict whether a stop is needed is redundant. The user confirmations from the application already provide this information definitively. D: Reinforcement learning is an overly complex approach for a problem with known, deterministic inputs. A deterministic optimization algorithm is far more efficient and suitable for this VRP scenario.

References

1. Google Developers, OR-Tools Documentation: The Vehicle Routing Problem (VRP) is defined as finding the optimal set of routes for a fleet of vehicles to service a given set of customers (or stops). The documentation states, "The input to a VRP is a set of locations..." In this scenario, the confirmed shuttle stations are the known locations, making it a direct application for an optimization solver like OR-Tools, not a predictive model.

Source: Google Developers, OR-Tools, "Vehicle Routing". (https://developers.google.com/optimization/routing/vrp)

2. Stanford University, CS221 Course Notes: Artificial Intelligence courseware distinguishes between problem types. This scenario falls under "Search and Optimization" problems, where the goal is to find an optimal plan given a known model of the world (the confirmed stops and map), rather than "Supervised Learning," which involves making predictions based on past data.

Source: Stanford University, CS221: Artificial Intelligence: Principles and Techniques, Autumn 2019, Lecture Notes on "Search." (http://web.stanford.edu/class/cs221/lectures/search.pdf, Section 1: "Search problems")

3. Google Cloud Solutions: Google's official solutions for logistics and routing problems emphasize optimization over prediction for this specific task. The "Route optimization" solution pattern uses services like Cloud Optimization AI (or OR-Tools) to solve VRPs, taking known locations and constraints as inputs to generate optimal routes.

Source: Google Cloud Architecture Center, "Route optimization: Reference architecture". (https://cloud.google.com/architecture/route-optimization-reference-architecture)

Keep practicing

Take the free PROFESSIONAL-MACHINE-LEARNING-ENGINEER practice test

Ten exam-style questions with answers and explanations, plus the exam facts and study guides.

Start the free practice test