ForgeGrid | Autonomous Production Intelligence

⚡ Technical Executive Summary

ForgeGrid is a real-time production intelligence system created for Hackathon dla Małopolski 2025 (Project FailSafe). The platform monitors 8 industrial machines and automatically manages task scheduling using the proprietary ForgeFlow™ MCT-S (Minimum Completion Time with Setup Awareness) algorithm.

🎯 Key Performance Metrics

Metric	Value	Context
Response Time	<200ms	From failure detection to new schedule generation
Loss Reduction	93%	Compared to manual planning
Annual ROI	$4,032,000	At 10 incidents/month
Savings per Incident	$33,600	28 minutes × $72k/hour
Machines Monitored	8	5× CNC, 2× Assembly, 1× Test
Task Types	12	Various production operations
Throughput	500+ tasks/h	Under optimal conditions
Memory Footprint	~50KB	Per 100 active tasks

🏭 Machine Fleet

ID	Machine	Type	Operations
M1	Haas VF-2	CNC	milling, drilling, threading, boring, facing, chamfering
M2	DMG MORI	CNC	milling, drilling, threading, boring, facing, chamfering
M5	Mazak	CNC	milling, drilling, threading, boring, facing, chamfering
M6	Okuma	CNC	milling, drilling, threading, boring, facing, chamfering
M8	Haas VF-4	CNC	milling, drilling, threading, boring, facing, chamfering
M3	Kuka Robot	Assembly	assembly, welding, wiring
M7	Fanuc Robot	Assembly	assembly, welding, wiring
M4	EOL Station	Test	testing, calibration, packaging

🔗 Links:

• Landing Page: https://netbr3ak.github.io/forge-page/
• Live Demo: https://netbr3ak.github.io/forge-grid/
• Main Repository: https://github.com/NetBr3ak/forge-grid

---

🧠 ForgeFlow™ MCT-S Algorithm

MCT-S = Minimum Completion Time with Setup Awareness

Theoretical Foundations

• Johnson's Rule - Optimal sequencing for 2-machine flow shop
• List Scheduling - Priority-based greedy assignment
• Work Stealing - Dynamic load balancing from parallel computing
• Bin Packing - LPT heuristic for makespan minimization

4-Phase Pipeline

1. Task Sorting: Priority → Constraint Tightness → Longest Processing Time (LPT)
2. Machine Scoring: Score = (ETA × priorityWeight) + setupPenalty + transportTime - preferenceBonus
3. Optimal Allocation: Select machine with lowest score, tie-breaker: prefer no setup change
4. Work Stealing: Idle machines scan busy queues for compatible tasks with positive benefit

Algorithm Parameters

Parameter	Value	Purpose
TRANSPORT_TIME	5 min	Inter-station transfer overhead
SETUP_TIME	3 min	Changeover penalty (different task type)
BATCH_WINDOW	5	Tasks analyzed simultaneously
REFRESH_RATE	300ms	System tick interval

Performance Metrics

Metric	Value
Scheduling Latency	<200ms
Rebalancing Time	<500ms
Work Steal Rate	~15%
Setup Optimization	~40% reduction

---

🏗️ Architectural Decisions

1. Zero-Dependency Vanilla JavaScript (IIFE Pattern)

• Decision: We rejected heavy frameworks (React/Vue) in favor of pure Vanilla JS wrapped in an Immediately Invoked Function Expression (IIFE).
• Why:
  • Performance: Eliminates bundle overhead. The entire JS payload is <5KB gzipped.
  • Encapsulation: The IIFE pattern creates a private scope, preventing global namespace pollution.
  • Maintainability: Clear separation of concerns without build step complexity.

2. Utility-First CSS (Tailwind via CDN + Custom Config)

• Decision: Utilized Tailwind CSS with a custom tailwind-config.js script.
• Why:
  • Rapid Prototyping: Immediate iteration on the "Industrial Cyber" aesthetic.
  • Consistency: Custom configuration enforces the strict color palette and typography.
  • Performance: GPU-optimized animations ensuring 60fps scrolling.

3. Semantic HTML5 & Accessibility (WCAG 2.1)

• Decision: Strict adherence to semantic markup and ARIA standards.
• Why:
  • Inclusivity: "Skip to content" links, proper aria-label, and heading hierarchy.
  • SEO: Semantic tags provide clear content structure for search engines.
  • Internationalization: Separate entry points with correct hreflang tags.

4. Performance Optimization Strategy

• Resource Loading:
  • Preload: Critical assets preloaded to minimize LCP.
  • Lazy Loading: Off-screen elements use IntersectionObserver.
  • Network Awareness: Video autoplay respects user data saver preferences.
• Rendering:
  • CSS Variables: Dynamic theming with minimal repaint cost.
  • Compositor-Only Animations: Restricted to transform and opacity.

---

🛠️ Project Structure

.
├── assets/
│   ├── css/
│   │   └── styles.css        # Global styles, animations, and overrides
│   ├── js/
│   │   ├── app.js            # Core logic (Modals, Video handling, Scroll spy)
│   │   └── tailwind-config.js # Design system configuration (Colors, Fonts)
│   ├── images/               # Optimized raster assets & QR codes
│   └── videos/               # Compressed video assets
├── index.html                # English entry point (EN) - Default
├── pl.html                   # Polish entry point (PL)
├── 404.html                  # Custom error page
├── robots.txt                # Crawler directives
├── sitemap.xml               # SEO sitemap
└── README.md                 # Technical documentation

---

🚀 Deployment & CI/CD

Hosted on GitHub Pages with global CDN for low-latency delivery.

• Cache Policy: Aggressive caching headers.
• HTTPS: Enforced for security and HTTP/2 support.

---

📈 Business Case (ELPLC Validation Data)

The landing page communicates ForgeGrid's validated value proposition:

Production Context

• Partner: ELPLC (industrial validation partner)
• Throughput: 600 units/hour
• Unit Value: $120
• Hourly Revenue: $72,000

Cost Comparison

Method	Response Time	Cost per Incident
Manual Planning	30 min	$36,000
ForgeGrid	2 min	$2,400
Savings	28 min	$33,600

ROI Calculation

Savings per incident:     $33,600
Incidents per month:      10
Monthly savings:          $336,000
Annual savings:           $4,032,000

---

🛠️ Application Tech Stack

Layer	Technology	Version	Purpose
Frontend	React	19	UI Framework
Language	TypeScript	5.x	Type Safety
State	Zustand	5.x	Lightweight State Management
Styling	Tailwind CSS	4.x	Utility-First CSS
Animations	Framer Motion	11.x	Smooth Transitions
Charts	Recharts	2.x	Data Visualization
Export	ExcelJS	4.x	Professional Reports
Build	Vite	6.x	Fast Build Tool

---

🏆 Hackathon dla Małopolski 2025

Projekt FailSafe - Real-Time Production Intelligence System

Team Members

Person	Role
Szymon Jędryczko	Machine Learning Engineer
Agata Syc	Chief Operating Officer
Dominik Kamiński	Software Developer
Franciszek Głąb	Security Engineer
Rafał Tomaszewski	QA Engineer
Mateusz Pawlik	Technical Support

---

© 2025 ForgeGrid Systems. Engineered for resilience.