Problem Statement
Jeep and UV drivers juggle boundary, diesel spikes, and irregular passenger loads with no record — disputes with operators and family budgeting both suffer. Fleet apps track drivers for operators; drivers won’t adopt surveillance. Many drive with no load and old phones.
Proposed Solution
One-tap trip logging: start route, tap passenger counter, enter diesel and boundary at day’s end. Weekly page shows net kita, best/worst days, and diesel per km. A passenger-facing fare card explains the current matrix in Tagalog. GPS is off by default and only on-device if enabled; export is via QR the driver chooses to share.
Target Market
- Primary users: Jeep, UV, trike drivers and small 1-2 unit operators.
- Deployers: TODA/JODA officers, driver cooperatives.
How AI Is Used
- On-device weekly narrative: Turns rows into “Lunes pinakamahina dahil ulan; diesel tumaas ₱40” without cloud.
- Fare explainer: LM drafts polite passenger scripts for matrix changes.
- Why AI: Drivers need stories, not spreadsheets; templates can’t cover varied weeks.
Independent Deployment Angle
- No fleet platform, no telco deal, no venue, no headset. Works offline, any phone.
- Driver owns and can delete data anytime — the adoption unlock.
- TODA can print QR-based weekly proofs without a server.
Demo Scenario (3 minutes for judges)
- Log a mock day: 3 trips, diesel ₱800, boundary ₱900 — all offline.
- Weekly kita page appears with net and best day highlighted.
- Toggle GPS off; show “location never leaves phone” indicator.
- Show passenger fare card in Tagalog for a matrix change.
- Close: “Ang datos, sa driver.”
30-Hour Build Plan
| Hours | Task | Role |
|---|---|---|
| 0–3 | Ride along notes, define trip schema | All |
| 3–9 | SQLite log + kita math, QR export | Backend |
| 3–12 | On-device summary prompts, fare explainer | AI lead |
| 10–20 | Big-button driver UI, passenger card | Frontend |
| 20–26 | Driver-phrasing test, battery check | All |
| 26–30 | Pitch, disclosure | PM / Pitch |
Suggested stack: Flutter + SQLite, on-device Tagalog summary TinyLM. Public data / APIs: Public LTFRB fare matrices; team-made mock trips.
Risks & Privacy Notes
- Location sensitivity: GPS off by default; if on, coordinates never upload. Demo uses a fake route only.
- Money disputes: Rows are editable with history; app never auto-reports to operators.
- IP: All copy team-written; no scraped operator data.
AI Use Disclosure (draft)
Uses a small on-device model for weekly narratives and fare scripts. AI coding assistants helped with Flutter forms. All figures in the demo are fictional and checked by the team.
Rubric Scoring
| Criterion | Weight | Score (1–5) | Why |
|---|---|---|---|
| Innovation & creativity | 25% | 3 | Logbooks exist; driver-owned + offline narrative is the twist |
| Problem–solution fit | 25% | 5 | Fits distrust-of-surveillance reality |
| AI application & value | 20% | 3 | Modest but helpful summaries |
| Feasibility | 15% | 5 | Very buildable |
| Impact & scalability | 15% | 3 | Needs TODA champions |
| Weighted total | 3.80 |