1. Overview
2. Purpose
3. Data Source
4. Modeling Approach
5. Schema Structure
   • Fact Tables
   • Dimension Tables
6. Design Goals
7. System Boundaries
8. Intended Use Cases
9. Technology Compatibility
10. Architectural Notes

This repository contains an OLAP / analytical data model designed for management-level reporting in a pharmaceutical manufacturing environment.

The data model is built to support:

• Trend analysis
• GMP compliance monitoring
• Executive and operational dashboards

All analytical data is sourced from an upstream OLTP Quality Management System (QMS).

The primary purpose of this system is decision support, not operational data entry.

Key Principle:

OLAP = answers, not transactions

This model enables managers and quality leaders to understand what is happening, why it is happening, and where risks are emerging.

• Read-only data ingestion from the OLTP QMS database
• No direct user input
• No transactional responsibilities
• No enforcement of business rules

The OLTP system remains the system of record.

The data warehouse follows a Star Schema / Snowflake Hybrid approach.

• Fact tables capture quality-related events
• Dimension tables provide business and temporal context
• Schema is optimized for analytical queries and aggregations

• FactInspection
• FactCalibration
• FactDeviation
• FactMaintenance

Each fact table represents a measurable quality event with timestamps and quantitative attributes.

• DimEquipment
• DimTime
• DimShift
• DimProduct
• DimInspector

Dimensions are designed to enable slicing, filtering, and grouping of facts across business perspectives.

• High-performance analytical queries
• Clear and explicit business semantics
• KPI-ready schema design
• Historical trend and pattern analysis
• Manager-friendly aggregations

• A reporting and analytics layer
• Optimized for read-heavy workloads
• Designed for dashboards and BI consumption

• ❌ A transactional system
• ❌ A business rule enforcement layer
• ❌ A source of truth

• Quality performance dashboards
• Deviation trend analysis
• Risk hotspot identification
• Preventive quality signal detection

This data model is technology-agnostic and can be consumed by:

• Power BI
• SQL Server Analysis Services (SSAS)
• Any standard BI or analytics platform

This repository conceptually depends on the OLTP QMS system,
but remains physically and logically isolated to ensure:

• Clear separation of concerns
• Independent scaling
• Analytical performance optimization