A comprehensive digital solution that pairs an offline-capable Progressive Web App (PWA) with a scalable REST API to digitize factory floor workflows. It tracks tight manufacturing tolerances and optimizes warehouse dispatch operations using an advanced bounded-window algorithm supporting strict FIFO, LIFO, and Manual selection methodologies.
Problem Statement
Tracking production and inventory in a technical manufacturing plant like a yarn or poly-thread factory creates a massive logistical challenge. Managing thousands of yarn bobbins with varying attributes (such as denier, color, and twists) requires extreme precision.
Key challenges included:
- Manual ledgers failing to capture specific qualitative variations, causing dispatch errors.
- Significant mathematical complexity in calculating exact combinations of bags to fulfill large bulk weight orders without heavily overshooting the target weight constraint.
- Lack of operational tracking, resulting in older stock remaining undispatched and degrading.
- Intermittent factory-floor WiFi causing web forms and data-entry tools to drop crucial production entries.
- Uniform access limits lacking, meaning simple workers, quality QA, and dispatch operators all shared overlapping authority levels, threatening data integrity.
Solution
We developed a powerful platform comprising two central parts: an Offline-capable Frontend (PWA) and a scalable REST Backend, bridging floor mechanics with enterprise reporting.
Production Logging: Workers directly input thread bag completion parameters on their mobile devices recording specific operators, quality checkers, colour, denier, and net weights. Service workers synchronize data seamlessly even through connection drops.
Bounded-Window Target Fulfillment: A custom algorithm calculates the optimal combination of physical stock required to meet a specific buyer's target weight. By sorting by FIFO (First-In-First-Out for standard rotation) or LIFO (Last-In-First-Out for fresh batches), the system sequentially selects bags. When the target weight is about to be crossed, it forms a bounded window of nearby candidates and surgically selects the single bag that yields the smallest weight overflow, bypassing others to guarantee mathematical precision. Manual bag selection is also supported for complete customization.
Dispatch & Scanning: Physical loading is handled via real-time barcode scanning. The PWA seamlessly utilizes built-in mobile cameras or physical USB-OTG barcode guns directly connected to the device, verifying each bag as it boards a truck.
Role-Based Security & Automation: A bespoke RBAC configuration binds custom user roles securely within JWT tokens, strictly mapping them to physical factory hierarchies. Furthermore, completed dispatches funnel into an automated cron-job queue, sending scheduled notifications directly to buyers.
Conclusion
The integration of an offline-first PWA alongside a heavy-duty REST API transformed the tracking process within the high-volume yarn production facility. This digital ecosystem effectively removed human calculations from warehouse organization and mathematically perfected dispatch processes utilizing the Bounded-Window selection algorithm whilst preserving strict hierarchical control through deep role-based access controls.
By treating the architecture generically, the project provides long-term extensibility for future enterprise or logistical integrations.
Similar Use Cases
- Agricultural Commodities Management (tracking grain attributes and precise FIFO weight aggregations)
- Extrusion Plant monitoring (plastic pallets / resins by color and viscosity)
- Cold Storage & Distribution warehousing prioritizing expiration tracking via mobile barcode guns
- Custom WIP (Work-In-Progress) components tracing before final assemblies
