UPS Project Ops Decision Support (v1)

A Python-based operations decision-support and reporting tool modeled after an internal enterprise analytics system built for an individual UPS Franchisee. The tool transforms synthesized shipment data into KPI tables, trend visualizations, and an executive-ready PDF report to support operational decision-making.

DISCLAIMER: All data in this repository is 100% synthetic.

What this tool does

• Ingests shipment-level operational data (CSV)
• Computes core operational KPIs (service level, delays, exceptions, throughput)
• Generates clean charts for trend and performance analysis
• Produces a multi-page, stakeholder-ready PDF report
• Packages insights, assumptions, and roadmap into a single deliverable

Problem This Tool Solves

Large-scale logistics operations generate extensive daily performance data, but that data is often reviewed in fragmented reports that make prioritization difficult. Decision-makers may see metrics but lack a structured way to interpret operational risk, compare sites, and identify where intervention is most urgent. This tool consolidates operational KPIs into a ranked, decision-oriented report that supports faster, more consistent operational oversight.

Output

• PDF report: FedEx_Project_Ops_Report.pdf
• Screenshots:

Executive Summary

KPI Table

Trend Chart

Tech stack

• Python
• pandas
• matplotlib
• ReportLab
• Dependencies: requirements.txt

Project Structure

fedex-project-ops-decision-support/
├── README.md
├── LICENSE
├── requirements.txt
├── reports/
│   └── FedEx_Project_Ops_Report.pdf
├── src/
│   ├── export_pdf.py   # report orchestration
│   ├── metrics.py      # KPI computation
│   ├── rules.py        # deterministic decision logic
│   └── insights.py     # optional narrative inputs
├── data/
│   └── sample/
│       └── shipments_sample.csv
└── docs/
    └── screenshots/

How to run

By default, the tool runs on a synthetic sample dataset included at
data/sample/shipments_sample.csv.

You may substitute your own CSV if it follows the same schema.

Input CSV Requirements

To generate a report using your own data, provide a CSV file with the following columns:

Column Name	Description
date	Date of operations (YYYY-MM-DD)
scenario	Operating scenario (e.g. NORMAL, PEAK, DISRUPTION, LABOR_SHORTAGE)
package_volume	Total packages processed that day
network_capacity	Maximum network capacity for the day
on_time_rate	On-time delivery rate (0–1)
exceptions	Number of delivery exceptions
labor_hours	Total labor hours worked
cost_per_package	Average cost per package

Additional columns may be present but are ignored. You may use the provided sample dataset as a template.

Environment setup

python -m venv .venv
source .venv/bin/activate  # Windows: .venv\Scripts\activate
pip install -r requirements.txt
2. Generate the report
bash
Copy code
python src/export_pdf.py
The script outputs a PDF report to the reports/ directory.

Data

Sample data mirrors realistic operational patterns and distributions

Assumptions & limitations
Metrics are illustrative and simplified for demonstration purposes

Root-cause attribution is not automated in v1

Designed as a reporting artifact, not a live dashboard

Roadmap
Parameterized report configuration (date ranges, thresholds)

Automated anomaly flagging

Lane- and region-level drilldowns

Unit tests and CI workflow