Hourly Count vs Target Board

The problem this kills

By the time the day-shift numbers land on someone's desk, the shift is over and the gap is baked in. Nobody could react, because nobody could see the line falling behind hour by hour. The whiteboard at the end of the line gets wiped between shifts, the miss reasons live in a supervisor's memory, and the "why were we 400 units short?" conversation happens a day too late.

An hourly count vs target board fixes the timing. Each hour the operator writes down what actually came off the line against what was supposed to. The moment the line slips behind, everyone standing at the board can see it - and the reason gets captured while it's still fresh, not reconstructed from memory at end of shift.

What you'll build

A simple, login-protected web board for one or many lines:

• Targets in. Load each line/machine's hourly target rate for the shift from a CSV or Google Sheet (your existing planning spreadsheet works fine).
• Hourly entry. The operator enters actual good count for the hour. If it's under target, they pick a short reason from a controlled list (no free-text guessing).
• Live ahead/behind. The board shows this hour vs target, the running cumulative vs target, and a clear ahead/behind signal so a slipping shift is obvious at a glance.
• End-of-shift approval. The supervisor reviews the whole board and the miss reasons, then approves the shift before any of it counts as official.
• Export. One clean hourly-log CSV in the exact columns your daily report or OEE workbook expects.

What's inside the Implementation Plan

The plan is a step-by-step runbook you paste into an AI coding agent (Claude Code). It writes the actual app for you while you answer questions in plain language.

It opens by interviewing you about your plant - your lines and machines, how you name them, your shift pattern and hour boundaries, where target rates come from today, your real miss-reason categories, and your messy edge cases (partial hours, changeovers, a line that runs two products in one shift). Then it reflects a short tailored spec back to you and waits for your thumbs-up. The board it builds matches your floor, not a generic template.

From there it walks you through standing up the database, loading targets, the operator entry screen, the live board, the supervisor approval gate, and the export - each step ending in a ready-to-paste prompt.

The governance it includes (this is the point)

This isn't a throwaway spreadsheet. The plan bakes in the controls a real manufacturing record needs:

• Login so only your team can open the board.
• Row-level security so each plant/organization only ever sees its own lines and shifts.
• A full audit trail - who entered which count, when, and who approved the shift.
• A human approval gate - the hourly board is a draft until the supervisor reviews the numbers and miss reasons and approves the shift. Only then does it roll up to reporting.
• Duplicate guards - the same line + shift + hour can't be logged twice, so a double-entry never quietly inflates your numbers.

Who it's for

Operators who log the count, line leaders who watch the trend, and supervisors who own the shift number and the end-of-shift sign-off. It ties naturally into OEE performance-loss tracking - the miss reasons you capture here are exactly the data your OEE story is missing.

You've got this - paste the first prompt and let the agent interview you.