A maintenance dashboard for production lines that runs entirely in the browser: line health, a real ticket workflow, trend charts and data export. No backend, no build step, no dependencies — one HTML file, one stylesheet, one script.
The tools that actually get used on a shopfloor are the ones that work in
three seconds on whatever PC is next to the line. That constraint drives
everything here: localStorage instead of a server, <dialog> instead
of a modal library, hand-rolled SVG instead of a charting framework.
Dashboard — health / risk / open tickets per line with a rough "cost if this line stops" estimate, plus a searchable event log. Tapping a line opens a pre-filled ticket for it.
Tickets — a real lifecycle, not a counter: open (with line, priority, assignee), work, close. Closing requires a resolution note, because a closed ticket with no explanation is worthless three weeks later. Filter by state, export everything as CSV (RFC 4180 escaping included).
Trends — every "shift update" snapshots line health; the Trends tab draws the last 24 snapshots as an SVG line chart, plus cumulative opened-vs-closed ticket throughput. The charts are ~40 lines of code total; a charting library would be a hundred times that for two graphs.
Simulation with consequences — the shift update doesn't just move numbers: if a line's risk crosses 70% and has no open ticket, one is created automatically and logged, which is what a real condition monitor would do.
Data ownership — export the full state as JSON, import it on another machine, or reset to demo data. The storage schema is versioned and old installs migrate instead of breaking.
All demo data is invented — lines, people, numbers.
Open index.html in a browser. That's it. If your browser is picky about
file:// URLs:
python3 -m http.server 8080MIT — see LICENSE.