Build the application

## Build
Extract container drayage orders from vernacular emails and voice transcriptions across English, Malay, and Mandarin. Match requests against terminal availability and assign carrier jobs.

## User
Drayage fleet operations planner

## Workflow
1. Ingest hauling request transcripts submitted via chat applications or voice transcription logs.
2. Extract pickup locations, container numbers, delivery deadlines, and special equipment needs.
3. Cross-reference live port traffic reports to assign optimal Jurong or Keppel terminal collection slots.
4. Generate multilingual hauler dispatch itineraries and automated confirmation receipts.

## Interface
Operations map and dispatch board displaying active hauler locations, terminal congestion indicators, and pending booking cards.

## Data
Generate synthetic booking transcripts in three languages combined with public traffic feeds from data.gov.sg referenced in source 0.
Starter examples below are synthetic. They are not downloaded records.

## Source
[Singapore open data API](https://guide.data.gov.sg/developer-guide/api-overview)
Utilizes data.gov.sg Real-Time APIs for traffic and weather data to calculate transit buffers.

## Tools
langgraph, pydantic, fastapi.
LangGraph coordinates state transitions between speech entity parsing, route buffer calculations, and hauler dispatch.
Multilingual model with high precision in Asian regional dialects and entity extraction via API.

## Deliverables
Submit runnable source, pinned dependencies, setup commands, a working interface, example outputs, and a recorded end-to-end demo. Include tests and an action log.

## Acceptance
- Normalizes multilingual booking transcripts into structured container routing tickets.
- Incorporates real-time traffic delay buffers when assigning delivery windows.
- Flags chassis constraints such as side-loader requirements for unequipped vehicles.
- Generates clean driver dispatch sheets in the driver's preferred language.

## Starter Examples
```json
[
  {
    "input": {
      "transcript_lang": "ms",
      "raw_text": "Tolong hantar kontena MSKU1234567 dari Keppel Distripark ke Tuas South sebelum jam 3 petang, perlu kren.",
      "fleet_available": [
        {
          "driver_id": "D1",
          "equipment": [
            "crane_truck"
          ],
          "language": "ms"
        }
      ]
    },
    "expected_output": {
      "booking_valid": true,
      "assigned_driver": "D1",
      "container_no": "MSKU1234567",
      "pickup": "Keppel Distripark",
      "dropoff": "Tuas South",
      "driver_instruction_text": "Sila ambil kontena MSKU1234567 di Keppel Distripark sebelum 3 petang guna lori kren."
    }
  },
  {
    "input": {
      "transcript_lang": "zh",
      "raw_text": "集装箱明天要，哪个码头都可以，尽快联系我",
      "fleet_available": [
        {
          "driver_id": "D2",
          "equipment": [
            "standard_chassis"
          ],
          "language": "zh"
        }
      ]
    },
    "expected_output": {
      "booking_valid": false,
      "assigned_driver": null,
      "container_no": null,
      "pickup": null,
      "dropoff": null,
      "driver_instruction_text": null
    }
  }
]
```

## Safety
Generates simulated hauler assignments without communicating directly with live road vehicles or executing billing.
Use mock external writes. Require human review before any real deployment.