🚚 SmartCargo - Delivery Optimization in Romania

The transport company "SmartCargo România" needs intelligent solutions to improve delivery time estimates. You are the new data science specialist on the team, and your mission is to build accurate models that predict delivery times between cities in Romania.

🎯 Your main goal

You need to analyze historical data about trips between cities and understand how factors such as distance, time of day, weather, traffic, or driver experience influence the actual delivery duration.

Your goal is to build a Machine Learning model capable of predicting the estimated time for new trips.

---

📦 Transport data

Each row in the train_data.csv and test_data.csv files represents a delivery between two cities in Romania.

Each delivery contains the following details:

Column	Description
ID	Unique trip identifier
City A	Departure city (as text)
City B	Destination city (as text)
Distance	Actual distance between cities (in kilometers)
Time of Day	Time of day expressed in minutes from midnight when a trip departs (0 - 1439)
Weather	Weather during the trip (Clear, Rain, Snow, Fog)
Traffic	Traffic level on a numerical scale (0.0 - 1000.0), maximum means most congested
Road Quality	Road quality on a numerical scale (1 - 1000), maximum means highest quality
Driver Experience	Driver's experience level (1 - 40 years)
deliver_time	Only in train_data.csv: actual delivery time in minutes

---

🎯 Your mission

The operations manager has given you the following two essential tasks for improving "SmartCargo" operations.

The Prediction Dataset (test_data.csv) contains samples with the same characteristics as the training set, but without the predicted_time column.

Your model will generate predictions for these samples.

Subtasks

• Task 1 (20p): The Bârlad Situation: an important client from Bârlad has reported frequent delays on trips in fog conditions. The manager wants to find out how many trips depart from Bârlad and in fog weather (Fog) in the prediction dataset. Find and report the number of these trips.
• Task 2: Time Estimation for unknown trips: The manager has a list of new trips for which delivery times are unknown. You need to use historical data (from the training set) to train prediction models and then estimate delivery times for all trips in the prediction dataset.

Output format

A csv file output.csv that should include the following 3 columns:

• subtaskID - represents the subtask number (1, 2)
• datapointID - which refers to the ID column from test_data.csv
• answer - the corresponding answer for the datapoint for that subtask

Note: For subtask 1, which requires a single answer for the entire dataset, display a single line whose datapointID is 1.

Submit a single csv that contains the answers for all subtasks you have solved. To see an example, download the sample_output.csv file (Note: this, although in the correct format, gets 0 points on submission).

Score

Scores for subtasks will be calculated as follows:

• Subtask 1: integer answer, you get 20 points only if it is correct.
• Subtask 2: based on your model's performance, using the MAE (Mean Absolute Error) metric on the test dataset. You can find the exact method for calculating the score based on MAE in the Starter Kit.

👉 📦 Download Starter Kit OJIA 2025