← All case studies
PT Efisiensi Putra Utama (EPU)Transportation & Fleet OperationsActive — Phase 2 in progress (2025–ongoing)

AI Fleet Monitoring for an Indonesian Bus Operator

PT Efisiensi Putra Utama runs an intercity bus operation with 60+ units across Java. After delivering their ticketing platform in Phase 1, we are now building an AI fleet monitoring system that replaces 8 manual passenger checkers with CCTV-based counting and real-time bus tracking. Full ownership — from edge inference to operations dashboard. Real AI in real production, not a demo.

Client

PT Efisiensi Putra Utama (EPU)

The Problem

Phase 1 was a ticketing platform — seat selection, real-time tracking, and membership — which we delivered to production. Phase 2 is harder. EPU needed a way to verify passenger counts against manifests without putting 8 checkers at terminals every day, plus a live operational view of where each bus is and whether it is on schedule. The brief was practical: cut headcount cost in one specific role, give the control room a single map, and do it without ripping out the existing CCTV hardware. No "AI strategy" theater — a working system that survives Monday morning.

Constraints

  • Existing WVP-PRO + GB/28181 NVR hardware had to stay — no rip-and-replace
  • Edge inference required: bandwidth from buses is unreliable
  • Indonesian operations team — UI and SOPs must be in Bahasa
  • Must run on commodity infrastructure, not hyperscaler GPU clusters
  • Compliance with Hino telemetry API contracts and refresh-token lifecycle
  • Single-team delivery — no room for a 10-person AI lab

Our Approach

We took full engineering ownership end-to-end. Hardware integration with the existing WVP-PRO + GB/28181 NVR stack, snapshot pipeline into a YOLO26 inference sidecar, passenger count reconciliation against the digital manifest from Phase 1, and a Hino API client with token lifecycle management for GPS telemetry. On top of that — a real-time operations dashboard built on MapLibre + deck.gl with PostGIS-backed route geometry and OSRM polyline rebuilds for cleaner playback. The system is deployed on a multi-tenant Docker + Hetzner setup with Grafana, Prometheus, and Loki for observability. Clean Architecture in Go on the backend — domain, usecase, adapter, handler — so the modules survive past this project.

Tech Stack

Next.js 15Tailwind v4Go (Chi)PythonPostgreSQLPostGISMapLibredeck.glMQTTYOLO26OSRMDockerHetznerGrafanaPrometheusLoki

Outcome

  • Phase 1 ticketing platform live in production across the EPU fleet
  • AI passenger counting pipeline running on existing NVR hardware — no new CCTV needed
  • Real-time map view consolidating GPS, status, and manifest data into one operations screen
  • Replaces 8 manual checker positions when fully rolled out
  • Foundation for future fleet intelligence products under Idin Studio
  • Re-engaged organically by the client — Phase 2 awarded without a competitive pitch

Why this matters

Most "AI deployments" in Indonesia stop at the slide deck. This one runs at terminals every day, reconciles against real manifests, and is measured against a headcount line on the operations budget. That is the work we want. EPU is also a reminder of why we exist as an agency. The client came back to us directly — not through a marketplace, not through a sales funnel. Organic trust earned by shipping is the only acquisition channel that scales without burning capital.

LET'S GO

Response within 24 hours · Free discovery call

Ready to Start Your Project?

Tell us your idea or business challenge. We'll help determine the best next step — no upfront consultation fee, no pressure.

Start the Conversation
Top Rated Plus
100% Job Success
8+ Clients
AI Fleet Monitoring for an Indonesian Bus Operator | Idin Studio | Idin Studio