Optimizing Model Development Speed with BigQuery ML Regression

PMLE By islever Editorial Last updated

Developing ML models using BigQuery ML or AutoML on Gemini Enterprise Agent Platform. Building models given the task considering cost, complexity, latency, and scalability.

Short answer

A Create a BigQuery table and use BigQuery ML to build a regression model to maximize development speed for this small dataset.

Question

You work for a hospital that wants to optimize how it schedules operations. You need to create a model that uses the relationship between the number of surgeries scheduled and beds used. You want to predict how many beds will be needed for patients each day in advance based on the scheduled surgeries. You have one year of data for the hospital organized in 365 rows.
The data includes the following variables for each day:
• Number of scheduled surgeries
• Number of beds occupied
• Date
You want to maximize the speed of model development and testing. What should you do?
  1. Create a BigQuery table. Use BigQuery ML to build a regression model, with number of beds as the target variable, and number of scheduled surgeries and date features (such as day of week) as the predictors. Correct Answer
  2. Create a BigQuery table. Use BigQuery ML to build an ARIMA model, with number of beds as the target variable, and date as the time variable.
  3. Create a Vertex AI tabular dataset. Train an AutoML regression model, with number of beds as the target variable, and number of scheduled minor surgeries and date features (such as day of the week) as the predictors.
  4. Create a Vertex AI tabular dataset. Train a Vertex AI AutoML Forecasting model, with number of beds as the target variable, number of scheduled surgeries as a covariate and date as the time variable.

Key Takeaways

The main trap is selecting advanced AutoML Forecasting tools for simple regression tasks; however, dataset size and development speed requirements make BigQuery ML the superior, efficient solution.

Learn why BigQuery ML regression is the optimal choice for maximizing development speed when predicting bed occupancy from scheduled surgeries using limited historical data.

Common mistake: Candidates often choose Option D (Vertex AI AutoML Forecasting) because they recognize the time-based nature of the data. However, this overlooks the 'speed' constraint and the fact that the dataset is too small (365 rows) to justify the complexity and setup time of Vertex AI.

Expert Analysis

The question asks to predict a continuous value (number of beds) using a predictor variable (scheduled surgeries) and date features, with the explicit constraint to "maximize the speed of model development and testing". This is a classic regression problem, not a time-series forecasting problem requiring complex temporal modeling like ARIMA or AutoML Forecasting.

Option A is the correct choice because BigQuery ML allows for rapid model creation using standard SQL. With only 365 rows of data, the dataset is too small to benefit from the overhead of Vertex AI AutoML platforms (Options C and D), which are designed for larger datasets and more complex feature engineering. Furthermore, the prompt emphasizes speed; writing a few lines of SQL in BigQuery is significantly faster than setting up Vertex AI pipelines.

The most attractive but wrong answer is Option D (Vertex AI AutoML Forecasting). It sounds sophisticated and handles time-series data well, including covariates. However, it fails on two counts: first, it over-engineers a simple linear relationship problem; second, and more critically, it violates the "speed" constraint due to the setup and training overhead associated with Vertex AI services compared to the immediate execution environment of BigQuery ML. Comments highlight that while D is technically capable, it ignores the efficiency requirement and dataset size constraints.

Exam Strategy

Always look for keywords like 'speed', 'simple', or small dataset sizes. When these appear, prefer managed services with lower overhead like BigQuery ML over full-featured ML platforms like Vertex AI unless the problem complexity strictly requires them. Differentiate between true time-series forecasting needs and simple predictive relationships where regression suffices.

Frequently Asked Questions

Why not use ARIMA for daily bed occupancy?

ARIMA models are designed for univariate time series analysis. The question requires using 'number of scheduled surgeries' as a predictor, which makes this a multivariate regression problem, not a univariate forecast.

Is 365 rows enough for BigQuery ML?

Yes, for simple linear regression models, BigQuery ML can handle small datasets efficiently. AutoML platforms typically require much larger volumes of data to achieve meaningful accuracy improvements over simpler models.

More PMLE FAQ →

Community Votes

D
48%
A
28%
B
19%
C
5%

48% of anonymous learners picked answer D. Votes are pick records left by other test-takers — they are not the verified answer.

Discussion 30 contributions

Compiled from public community threads and reader submissions, lightly edited for clarity. Handles and vote tallies are shown as recorded, and comments reflect the views of individual contributors.

From public community threads 30

  1. lunalongo 👍 5 Answered A
    A is the best option because: - BigQuery ML allows model build/training within BigQuery using SQL - This is a regression model, not a timeseries forecast; no ARIMA (B) fit! - Data transfer to Vertex AI (C, D) and usage of AutoML not needed - AutoML is better for larger datasets, BigQuery ML works for 365 rows
  2. fitri001 👍 4
    A. BigQuery ML regression: While BigQuery ML offers quick model building, a regression model might not capture the time-series aspect of daily bed occupancy. Daily bed occupancy might have trends or seasonality which a plain regression model wouldn't capture. B. BigQuery ML ARIMA: ARIMA models are specifically for stationary time series data, and hospital bed occupancy might not always be stationary (e.g., holiday season might lead to higher occupancy). Additionally, ARIMA models typically don't incorporate additional features like the number of scheduled surgeries. C. Vertex AI AutoML Regression: Similar to option A, a regression model might not capture the time series aspect. While Vertex AI offers AutoML regression, using a solution designed for time series forecasting is more suitable here.
  3. b1a8fae 👍 4 Answered A
    A. Using BigQuery to comply requirement of speed of development. ARIMA does not use number of scheduled surgeries, and it is stated that the prediction must be based on that variable. So it must be A. LR model on BQ using scheduled surgeries, day of the week, etc, as predictors.
  4. forport 👍 4 Answered D
    'Vertex AI AutoML Forecasting' == for forecasting time series data
  5. VinaoSilva 👍 3 Answered D
    "You want to predict how many beds will be needed for patients each day" = Forecasting
Show 25 more comments
  1. shadz10 👍 3
    365 days of data may be insufficient for big query I’m going with C
  2. shadz10 👍 2
    While ARIMA models are commonly used for time series forecasting, they are more suitable for univariate time series data and might require additional manual intervention for feature engineering. In this case, we have multiple variables such as the number of scheduled surgeries, the number of beds occupied, and the date. AutoML Forecasting in option D is designed to handle multivariate time series data, and it automates much of the modeling process, including feature selection and hyperparameter tuning. This can potentially result in a faster and more efficient development and testing process compared to manually implementing and tuning an ARIMA model.
  3. fitri001 👍 2 Answered D
    Vertex AI AutoML Forecasting: This option leverages Vertex AI's AutoML capabilities for time series forecasting. It automatically explores different model types and hyperparameters to find the best fit for your data. This can significantly speed up model development compared to building a model from scratch. Date as time variable, surgeries as covariate: This approach acknowledges the time-series nature of bed occupancy with "date" as the time series variable. It also incorporates the "number of scheduled surgeries" as a covariate, allowing the model to learn the relationship between surgeries and bed usage.
  4. shadz10 👍 2
    D* not C Forecasting models are well-suited for predicting future values based on past trends, making them ideal for the goal of predicting bed occupancy for upcoming days. Dataset is too small for bigquery.
  5. Yan_X 👍 2 Answered D
    D, as B doesn't mention the 'number of scheduled surgeries'.
  6. shadz10 👍 2 Answered D
    D is correct I believe
  7. taylz876 👍 1 Answered A
    Why I lean A • The question's priority is speed. "Maximize the speed of model development and testing" is the deciding phrase. BigQuery ML regression is a few lines of SQL, while AutoML Forecasting is a longer, heavier training job, and 365 rows is far too little data to need it. • The question frames a relationship, not a forecast. It says to use the relationship between scheduled surgeries and beds, and the surgeries are known in advance. Predicting beds from tomorrow's scheduled surgeries is a regression on a known input. It isn't forecasting beds from past beds alone. • A handles the main time effect. The answer adds day-of-week features, which capture the weekly pattern hospitals have. The commenters who say regression can't handle time are overlooking that. • C is the trap. It says "minor surgeries", which isn't in the data. Where D has a point Hospital beds do carry over from day to day, since patients stay for several days, so a true time series model could capture that. Some commenters also note that BigQuery's ARIMA can take extra variables (ARIMA_PLUS_XREG in your notes), but B only uses date as the input, so it ignores surgeries, and that's why B is out either way.
  8. NamitSehgal 👍 1 Answered D
    A (BigQuery ML Regression): A simple regression model is not designed for time series forecasting. It wouldn't capture the temporal dependencies in the data and wouldn't be able to effectively predict future bed usage based on past trends. D Forecasting Task
  9. Fer660 👍 1 Answered D
    Not A or C: regression wil not do the trick because this is a time series. Not B: ARIMA is univariate, so if you are using the date as predictor, you won't be able to use the number of scheduled surgeries as predictor, and lose the key piece of information.
  10. info_appsatori 👍 1 Answered A
    IDK, i going with A, because its maximize the speed of development and testing. Also in question it says: You need to create a model that uses the """relationship"" between the number of surgeries scheduled and beds used. = linear regression problem.
  11. b2aaace 👍 1 Answered C
    I don't think this is a time series forecasting problem. The question clearly states that we should predict the number of beds based on the number of scheduled surgeries. this is a simple linear regression problem.
  12. pinimichele01 👍 1
    not b: ARIMA does not use number of scheduled surgeries, and it is stated that the prediction must be based on that variable
  13. pinimichele01 👍 1
    "You want to predict how many beds will be needed for patients each day in advance based on the scheduled surgeries."
  14. el_vampiro 👍 1 Answered B
    BQML ARIMA+ is the quickest to develop and test as it trains faster than AutoML. It can do Multivariate forecasting.
  15. Hrishikesh1992 👍 1 Answered B
    We have only 360 rows of data, I went with it because we require statistical model here rather than ML models.
  16. el_vampiro 👍 1
    ARIMA can do multivariate and also take holidays into account. Also trains much faster than AutoML
  17. samantha84 👍 1 Answered A
    There aren't enough data for AUTOML Forecasting. In this context, the option A is correct.
  18. 8c200dd 👍 1 Answered A
    in question they said, minimize the speed of model development and tesing. so A
  19. shadz10 👍 1
    https://cloud.google.com/vertex-ai/docs/tabular-data/forecasting/overview
  20. sonicclasps 👍 1 Answered D
    best suited, and treats the input as a time series, unlike A
  21. kalle_balle 👍 1 Answered B
    Using B instead of D as it requires speed of development.
  22. dija123 👍 1 Answered D
    Train a Vertex AI AutoML Forecasting model
  23. el_vampiro 👍 1
    ARIMA can do multivariate timeseries
  24. CHARLIE2108 👍 1 Answered B
    I went with B.
  25. pinimichele01 👍 1 Answered D
    best suited

Reader comments & corrections

No reader corrections yet — spotted an error or have a note? Share it below.

Log in to comment, report an error, or add a note about this question.

Monitoring ML Model Performance with Precision and Recall Sampling

S-grade

Optimizing ML Model Recall for Compliance Risk

S-grade

How to Minimize Development Time for Weekly TensorFlow Retraining in Vertex AI?

S-grade

Building CLV Models with BigQuery ML in Vertex AI Workbench

S-grade

Practice All PMLE Questions

Access 65 questions with complete answers and detailed explanations.

View Full PMLE Practice Test →

← Back to PMLE Study Guide