Bionics

Solution Overview

Bionics, in strategic partnership with ProdataKey (PDK), has developed a real-time, cloud-native analytics platform that transforms employee badge-swipe data into actionable insights. Leveraging the Microsoft Azure ecosystem and the Power Platform, this solution enables firms to visualize, analyse, and respond to workplace patterns with precision and agility. 

Designed for modern, hybrid work environments, the platform ingests badge-swipe data from PDK APIs every minute, stores it securely in Azure Data Lake Storage (ADLS), and processes it through Azure Databricks and Azure Data Factory for enrichment and transformation. Curated data is then made accessible through Azure SQL Database and visualized using Power BI, enabling enterprise-wide transparency and decision-making. 

Business Impact and Strategic Value 

With hybrid work models becoming the standard, organizations often lack real-time visibility into how and when office space is being used. This solution addresses that gap by equipping executive, HR, and real estate teams with the insights they need to: 

• Monitor live and historical trends in office attendance 
• Identify peak and off-peak usage patterns across locations 
• Make data-driven decisions on office consolidation, expansion, or redesign 
• Enforce and track compliance with in-office work policies 
• Provide employees with greater clarity and planning tools around on-site expectations 

This end-to-end solution not only improves operational efficiency but also supports strategic planning, workplace optimization, and policy adherence in a hybrid work environment. 

Technical Architecture Overview 

The solution is architected to support a near real-time data ingestion and analytics pipeline, enabling continuous integration of employee access data from ProdataKey (PDK) into enterprise-grade business intelligence dashboards. The design emphasizes modularity, scalability, security, and reuse, supporting both single-tenant and multi-tenant client deployments. Below is a breakdown of the key architectural components: 

End-to-End Architecture Components 

1. ProdataKey API (Source System) 

• Provides badge-swipe event data at one-minute intervals using a secured RESTful API. 
• Acts as the authoritative source of physical access control data (door entries, swipes, timestamps, device IDs). 
• Authentication handled via secure tokens or API keys, integrated with Azure secrets for secure credential storage. 
• Flexible to accommodate extensions like door status, location hierarchy, or alarm events from PDK's broader API suite. 

2. Azure Databricks (Streaming and Data Ingestion Layer) 

• Hosts Python notebooks that are scheduled or triggered to hit the PDK API at predefined intervals (e.g., every minute). 
• Leverages the Databricks runtime for secure, scalable ingestion of real-time access data. 
• Stores the raw output in Azure Data Lake Storage in a structured format (Parquet/Delta). 
• Built-in monitoring and logging for API connectivity issues, ingestion failures, and retry logic. 
• Enables real-time anomaly detection (e.g., badge swipes after hours) through streaming logic extensions

 3. Azure Data Factory (ETL/ELT Orchestration) 

• Acts as the central orchestration engine for all data pipelines. 
• Triggers and manages: 
  • Scheduled ingestion notebooks (via Databricks or REST) 
  • Schema validation, normalization, and deduplication 
  • Batch transformation pipelines (daily, hourly) 
• Enriches raw data with: 
  • Office metadata (location, region) 
  • HR reference data (employee department, role) 
  • Calendar data (weekdays, holidays) 
• Implements data quality rules and business logic (e.g., identify first swipe per user per day). 

4. Azure Data Lake Storage (Raw Data Layer) 

• Serves as the central data lake for multi-layered data staging: 
  • Bronze Layer: Raw PDK API output (immutable) 
  • Silver Layer: Validated and enriched data 
  • Gold Layer: Curated, analytics-ready datasets 
• Supports fine-grained access control via Azure RBAC and ACLs. 
• Provides historical storage with auditability for regulatory compliance or usage analytics. 

5. Azure SQL Database (Curated Serving Layer) 

• Stores transformed, business-ready datasets for reporting and semantic modeling. 
• Acts as the primary data mart, exposing fact and dimension tables such as: 
  • FactBadgeSwipe 
  • DimEmployee 
  • DimLocation 
  • DimDate 
• Optimized for direct query and scheduled refresh in Power BI. 
• Implements row-level security (RLS) for client/user-specific data access. 

6. Power BI Semantic Model 

• Developed using Power BI Desktop, connecting directly to Azure SQL. 
• Incorporates: 
  • DAX measures for swipe counts, unique users, occupancy rates 
  • Time intelligence (week-to-date, month-over-month trends) 
  • Calculated columns for business logic (e.g., “first entry of day”) 
• Enables real-time data refresh (DirectQuery) or scheduled import, based on performance needs.
• Supports composite models when needed (e.g., Azure SQL + ADLS). 

7. Power BI Service (Visualization & Distribution Layer) 

• Dashboards and reports are published to Power BI Service workspaces, enabling web/mobile access.
• Implements: 
  • Role-based access by department, geography, and management level 
  • Data alerts for anomalous patterns (e.g., no entries for a location by 11 AM) 
  • Scheduled refreshes aligned with leadership reporting cadences (daily, weekly) 
• Dashboards are embedded or shared directly via secure workspace links, Microsoft Teams, or email.
• Optionally integrates with Power Automate for workflow triggers based on usage thresholds or policy breaches. 

Security & Compliance Highlights: 

• End-to-end encryption: TLS in transit and Azure-managed encryption at rest. 
• Audit logging via Databricks, ADF, and Power BI Service. 
• GDPR and SOC2-aligned practices for access logging and data handling. 
• Segregation of data for multi-client environments using schema-level isolation and RBAC.

The architecture diagram below illustrates the end-to-end data flow and platform components described in this section, highlighting how badge-swipe data is ingested, processed, stored, and visualized across the Azure ecosystem

Figure 1: Architecture Diagram

Interactive Dashboards 

The PDK – Access Control Power BI report serves as the primary front-end for visualizing badge-swipe activity and related access analy cs. This business-facing interface is designed to support decision-making by HR, real estate, and executive teams. Developed using Power BI Service, it provides secure, role-based access to interactive reports. 

The following dashboards are included and fully customizable: 

• Office Visits Over Time: View daily counts of office entries by location 
• Card Swipe Volume: Analyse raw badge activity patterns 
• Top Visitors & Department Trends: Filter by role or department 
• Heatmaps & Utilization Charts: Know when and where traffic is highest 

Figure 2: Employee Visit Patterns and Badge Usage

Figure 3: Workplace Access Summary by Location and Hourly Patterns

Figure 4: Employee Access Patterns and Swipe Forecasting 

Client Testimonial - BRC CPA  

"Bionics delivered an incredibly intuitive and visually impressive solution. It gave us immediate clarity into office usage trends, helping us make data-backed decisions on space and policy. Their innovation and responsiveness made them the ideal analy cs partner."  

- COO, Bernard Robinson & Company (BRC CPA)