Do I need sensors on the machines?

Not to start. The schedule, completions and downtime log you already keep answer most of the morning meeting's questions.

How long does a first view take?

Usually a few weeks, most of it agreeing the delay reasons and cleaning the records.

A shop-floor view without buying a new MES

Plant managers want to see what is late, what is stopped and why, today. The usual answer is a manufacturing execution system and a long rollout. For many plants, a simple view built from records they already keep gets most of the way there in weeks.

Ask a plant manager what they want from software and the answer is often simple: show me, every morning, what is late, what is stopped and why. The usual industry answer is a manufacturing execution system, with machine connections, operator terminals and a rollout measured in quarters.

For a large plant that can be right. For a plant of fifty to three hundred people, a much smaller step usually gets most of the value first, using records the plant already keeps.

What you probably already have

Orders and due dates in the ERP.
A schedule, often in a spreadsheet, showing what runs on which line and when.
Job completions booked in the ERP, sometimes a day late.
Downtime notes, in a log, a spreadsheet or on paper at each machine.
Material receipts, in the ERP or in emails from purchasing.

Each of these is incomplete on its own. Joined together, they answer most of the morning meeting's questions.

8.3% of planned production time lost to downtime in US discrete manufacturing (D. S. Thomas, NIST, Annual Report on the U.S. Manufacturing Economy: 2025 (NIST AMS 100-76))
$245B a year, the cost of that downtime (D. S. Thomas, NIST, Annual Report on the U.S. Manufacturing Economy: 2025 (NIST AMS 100-76))

The view, in four parts

What is late Every open order whose due date has passed or is at risk, with the job and the line it is waiting on.
Why it is late A reason from a short agreed list: material, machine, labour, quality, customer change. Free text is allowed, but the list is what gets counted.
What stopped Downtime by machine for the last day and week, with the reason and how long.
The trend The same numbers by week, so the meeting can see whether things are getting better, not just how bad yesterday was.

Joined together, the records a plant already keeps answer most of the morning meeting's questions.

How it is built

Each night, a small job reads the ERP, the schedule spreadsheet and the downtime log, joins them on order and job numbers, and writes the result to a reporting database. The view reads from there, so nobody's working spreadsheet is ever locked or changed.

The hardest part is never the technology. It is agreeing the list of delay reasons with the people who record them, and making recording easy enough that it happens. We usually spend more time on that than on the code. A delay reason picked from a list on a tablet at the line beats a perfect taxonomy nobody uses.

Where AI helps

Mostly it does not need to. This is joining and counting records, and plain code does that perfectly. The one useful job is turning years of old free-text downtime notes into the new reason list, so the trend starts with history instead of from zero. A person checks the ones the model is unsure about.

When you do need an MES

If you need real-time machine data, traceability down to the lot for regulated customers, or electronic work instructions at every station, a proper MES is the right tool. Even then, the view described here is a useful first step: it shows which lines and causes matter most, which tells you where an MES should start, and it gets the team used to recording reasons consistently.

Questions people ask

Do I need sensors on the machines?: Not to start. The schedule, completions and downtime log you already keep answer most of the morning meeting's questions.
How long does a first view take?: Usually a few weeks, most of it agreeing the delay reasons and cleaning the records.

Annual Report on the U.S. Manufacturing Economy: 2025 (NIST AMS 100-76), D. S. Thomas, NIST (2026)

Want a morning view in weeks? Send us a description of where your schedule, completions and downtime live today. We will tell you what the first version could show.

Canonical URL: https://ugibo.com/blog/a-shop-floor-view-without-a-new-mes