Index / TyreSense / Sensor Tracker

Case Study · 01 / 03

Sensor Tracker

An internal web platform that lets TyreSense's manufacturing partner validate sensor boards at the source — allowing passing units to move directly toward production while failed or flagged units are routed back to the repair lab. Designed, built, and deployed solo during my Co-op Production Engineering term at TyreSense.

Role
Solo · Design → Deploy
Context
TyreSense · Co-op (VTS)
Status
Live · In production use
Scale
19,838+ sensors tested as of June 5th, 2026

Problem: All sensor boards were routed through VTS, turning the repair lab into a sorting bottleneck.

What I built: A partner-side validation platform with automated tests, firmware checks, and centralized logging.

Impact: Passing units now move toward LTS directly, while VTS focuses only on failed or flagged boards.

Source note: Estimated timing/volume figures below are derived from internal weekly throughput observations and process mapping, and remain labeled as estimates pending validated production reporting.

01

The problem

TyreSense's sensor boards are built by a manufacturing partner and shipped in for handling at VTS, the R&D and repair lab. Every unit — around 1,500 a week — arrived as a single mixed batch. VTS technicians tested each board by hand, manually repackaged the working units, then re-shipped them across to LTS, the production site.

That meant every sensor took a full inbound + outbound trip through VTS before it ever reached the line, regardless of whether it actually needed repair. The bottleneck was structural, not technical: the lab was being used as a sorting facility.

Sensor flow · before vs. after Where boards travel between leaving the partner and reaching production.
02

Approach

Move validation upstream, to the partner. If the same tests VTS runs can be run at source — with the same rigor, the same audit trail, and a single shared archive — then the lab stops being a sorting facility and becomes what it should be: a repair lab.

The system had to be opinionated about correctness (no ambiguous pass/fail), unforgiving about audit (every test timestamped and logged), and friendly enough for a technician to operate without a manual.

03

What I built

  • Automated test workflow

    End-to-end sensor validation with real-time pass/fail, run by the partner — no hardware round-trip.

  • Hardware diagnostics

    Battery level, voltage, lifespan, and RSSI signal-strength checks captured per device.

  • Firmware validation

    Version-aware checks that gate boards on the right firmware before they leave the line.

  • Centralized database

    Every result written with a timestamp. Long-term archive, searchable by anyone with access.

  • Failure-rate analytics

    Dashboard surfacing failure trends over time — turning raw test data into operational insight.

  • Role-aware reporting

    Searchable records for technicians, analysts, managers, and stakeholders. One source of truth.

04

Screens

Two views from the deployed platform. Click any image to expand.

Login · operator entry point
Dashboard · test results & failure trends
05

Stack

Frontend

  • Internal web UI
  • Operator-facing test interface

Backend

  • Software automation
  • Database-driven logging

Data

  • Timestamped test archive
  • Failure-rate analytics

Hardware

  • Sensor circuit board testing
  • Firmware validation
  • RSSI / voltage diagnostics
06

Outcome

19,838+ sensors tested since deploy as of June 5th, 2026
Live in production use
Failures only failed or flagged units routed to the lab

Passing sensors can now move from the manufacturing partner directly toward production at LTS. VTS is reserved for failed or flagged units that actually need repair work — turning the lab from a manual sorting checkpoint into a data-driven repair workflow.

Estimated time-to-production · partner → line Approximate calendar time for a passing sensor to reach the production floor.

Estimate source · internal weekly throughput observations + process mapping; pending validated production reporting.

Weekly volume routed through VTS Sensors requiring VTS handling each week.

Estimate source · internal weekly throughput observations + process mapping; pending validated production reporting.

Note · Some technical details remain confidential. Happy to walk through specifics in conversation.