LockerStation AI Hub is an AI-powered operations platform built for Pick!, Singapore’s islandwide parcel locker network. The project tackles a real last-mile logistics challenge: keeping hundreds of locker stations running reliably while planning where the network should grow next. Rather than treating maintenance and expansion as separate spreadsheets, the hub brings them into one intelligent admin console. Built as a modern web application with real AI integrations, it gives operations teams a live view of every station—from Marina Bay to Jurong East—along with tools to diagnose issues, manage capacity, and make data-driven expansion decisions. The platform is designed as a proof-of-concept demo that shows how artificial intelligence can move locker operators from reactive firefighting to proactive, strategic network management across residential hubs, malls, and transit locations.
Locker Management
The Locker Management module is the day-to-day control room for the entire fleet. An interactive map plots all locker stations across Singapore, while search and regional filters (Central, North-East, East, West, North) make it easy to find any site instantly. The station list shows operational status, address, and real-time availability—for example, “8/12 Free” at Marina Bay Locker Station. Selecting a station opens a compartment grid where each locker displays its size (S, M, L, XL), dimensions, and occupancy state via color-coded indicators. Administrators can add new stations, remove underused ones, and drill into individual compartments to run AI diagnostics on sensor health and predictive maintenance needs. A one-click AI Strategic Network Planning action connects this module directly to expansion analysis, turning everyday locker oversight into a foundation for smarter network decisions.
AI Network Expansion
The AI Network Expansion module shifts from daily operations to long-range strategy. When triggered, multi-agent AI analyzes network saturation, historical demand, revenue patterns, partner performance, and maintenance signals to produce a comprehensive expansion report. The executive summary highlights strategic opportunities—such as high-growth zones like Punggol North, Tengah Town, and Woodlands South—alongside a network health score, revenue uplift estimates (up to 20%), and a 6–12 month demand forecast. Recommendations are organized into three actionable buckets: New Locations for greenfield deployment with priority ratings and estimated costs; Stations to Expand where existing high-utilization sites need additional L and XL compartments; and Relocation Candidates flagging underperforming stations like Sentosa Boardwalk or Novena Square with risk scores. Together, these insights help Pick! invest where demand is rising and optimize capacity where it already exists.
Technologies Used
- Platform: React 19 +Next js
- AI Models: Google Gemini
- Database: Firebase — Auth, Firestore, Cloud Storage
- Infrastructure: Vercel Edge Network
- Industries: Logistics & E commerce
