Shopee Challenge

General Discription

Due to the recent COVID-19 pandemic across the globe, many individuals are increasingly turning to online platforms like Shopee to purchase their daily necessities. This surge in online orders has placed a strain onto Shopee and our logistics providers but customer expectations on the timely delivery of their goods remain high. On-time delivery is arguably one of the most important factors of success in the eCommerce industry and now more than ever, we need to ensure the orders reach our buyers on time in order to build our users’ confidence in us.

In order to handle the millions of parcels that need to be delivered everyday, we have engaged multiple logistics providers across the region. Only the best logistics providers that are able to meet Shopee’s delivery standards are partnered with us.

The performance of these providers is monitored regularly and each provider is held accountable based on the Service Level Agreements (SLA). Late deliveries are flagged out and penalties are imposed on the providers to ensure they perform their utmost.

The consistent monitoring and process of holding our logistics providers accountable allows us to maintain our promise of timely deliveries to our buyers.

Task

Identify all the orders that are considered late depending on the Service Level Agreements (SLA) with our Logistics Provider.

For the purpose of this question, assume that all deliveries are considered successful by the second attempt.

For more detail about the challenge, please visit Kaggle

Note

This is my personal approach for Shopee Challenge. In this project, I try to utilize the chaining method in 'modin[pandas]' and refactor code as function (rather than split them in several code cell). This is a useful practice which helps us the avoid duplicate the code. Another approach is to use polars which is a dataframe library written entirely in Rust using the Appache Arrow file format.

Both approaches have their advantages. With modin[pandas] as a drop-in replacement, you can start to analyze your data without learning new API and the running time is also quite fast. On the other hands, you need to learn new API with polars (this is not so hard if you are familar with SQL) and the running time of polars is much more faster than modin[pandas], like crazy fast, about 40-50 times.

On the installation experience, modin[pandas] takes time to install because of its huge dependencies while polars just takes 5 seconds since it is written entirely on Rust.

Here is the link to download the data

Requirements:

• Basic library for scientific computing: pandas, numpy,modin[ray], polars

Notebook Organization

• analysis.ipynb: this notebook uses modin[ray] to complete the tasks.
• analysis_polars.ipynb: this notebook uses polars to complete the tasks.