Skip to content

With the growing usage of ecommerce websites, The demand to get the product delivered is also increasing. So to meet today's societies expectation to deliver the product as early as possible we need predictive analysis of shipments and why certain shipments are being delayed.

Notifications You must be signed in to change notification settings

aarshg28/Supply-Chain-Analysis

 
 

Repository files navigation

BigDataProject

Shipping Time Prediction Analysis

Project Problem Statement

With the growing usage of ecommerce websites, The demand to get the product delivered is also increasing. So to meet today's societies expectation to deliver the product as early as possible we need predictive analysis of shipments and why certain shipments are being delayed.

This dataset contains complete shipping data for all products delivered including mainly Country,Managed By,Fulfill Via,Vendor INCO Term,Shipment Mode,PQ First Sent to Client Date,PO Sent to Vendor Date,Scheduled Delivery Date,Delivered to Client Date,Delivery Recorded Date,Product Group,Sub Classification,Vendor,Item Description,Molecule/Test Type,Brand,Dosage,Dosage Form,Unit of Measure (Per Pack),Line Item Quantity,Line Item Value,Pack Price,Unit Price,Manufacturing Site,First Line Designation,Weight (Kilograms),Freight Cost (USD)

Overview:

  1. Led end-to-end ETL for supply chain health commodity shipment and pricing data, enhancing global spending insights.
  2. Conducted exploratory data analysis and dashboard creation for data distribution across countries, delivery types, and vendor pricing.
  3. Implemented data preparation, including date column conversion and feature engineering, optimizing for Amazon SageMaker.
  4. Utilized TensorFlow for analytics, achieving a training RMSE of 25.86 and a test RMSE of 24.73. Addressed data challenges, including impure values and missing data, mitigating bias through data cleaning and imputation.

Data Structure :

columns

ID Project Code PQ # PO / SO # ASN/DN # Country Managed By Fulfill Via Vendor INCO Term Shipment Mode PQ First Sent to Client Date PO Sent to Vendor Date Scheduled Delivery Date Delivered to Client Date Delivery Recorded Date Product Group Sub Classification Vendor Item Description Molecule/Test Type Brand Dosage Dosage Form Unit of Measure (Per Pack) Line Item Quantity Line Item Value Pack Price Unit Price Manufacturing Site First Line Designation Weight (Kilograms) Freight Cost (USD) Line Item Insurance (USD)

Data Set : https://www.usaid.gov/opengov/developer/datasets/SCMS_Delivery_History_Dataset_20150929.csv

About

With the growing usage of ecommerce websites, The demand to get the product delivered is also increasing. So to meet today's societies expectation to deliver the product as early as possible we need predictive analysis of shipments and why certain shipments are being delayed.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Jupyter Notebook 100.0%