A press dashboard is not a better run
# A press dashboard is not a better run The screen in the office was full, and the screen was pleased. Counters for the sheetfed press, a tile for the digital machine, a running total of sheets, a row of stops, a waste figure somebody had remembered to switch on. Green, mostly. The production lead
A press dashboard is not a better run
The screen in the office was full, and the screen was pleased. Counters for the sheetfed press, a tile for the digital machine, a running total of sheets, a row of stops, a waste figure somebody had remembered to switch on. Green, mostly. The production lead looked at it before the morning meeting and felt, for a minute, informed. Out in the hall the job that should have gone at six was still on a pallet by the guillotine. The lorry had left. The customer had already rung. Nothing on the pleased screen said "late", because late wasn't a counter. Late was a promise, and the promise lived on a ticket the dashboard didn't open.
Seeing the press is not the same as changing the next job. A dashboard is a view. A better run is a different choice, made in time, by someone allowed to make it. Plants buy the view because the view is what a vendor can put on a wall, and because a busy hall is hard to see without help. The help stops at the glass. If the next job is wrong, unapproved, on the wrong stock, or already impossible in finishing, a counter of the current impression will watch you find that out the slow way.
What the counters actually are
A counter counts. Sheets past a sensor. Clicks on a digital press. Stops, if the press defines a stop in a way you understand. Time in a state the software calls running, makeready, or idle. Waste, if a person or a sensor separated waste from good work, which is less common than the tile implies. These are facts about the recent past of a machine, and sometimes about the present. They are not facts about a customer's due time, except where you've done the extra work of tying the machine's count to a job and the job to a promise.
That tie is where most dashboards quietly fail. The press was "running" while the minder was pulling makeready sheets on the previous job number, because nobody had closed it. The digital press was "productive" on a queue that included reprints of work you'd already invoiced. The waste tile was calm because spoiled finishing never passed the press sensor at all. The board showed a busy plant. The accountant, later, showed a month that hadn't made what the busyness suggested. Both reports were accurate about different things. Only one of them was on the wall at the meeting.
You can hand a dashboard an honest pulse. A count you've checked against a pile. A job number changed when the job changes. A waste count that means waste. You can't hand it the pile it wasn't shown, the spoilage in the gluer, or the due time nobody entered. If the wall disagrees with the hall, the hall is the one printing. People will glance at the wall, feel the glance fail them twice, and stop using it. A dashboard nobody believes is furniture. It still costs you the time spent typing numbers into it so the furniture looks alive.
A view that arrives late, or flatters you
Even an honest counter can be useless for the decision in front of you. If the tiles update when the shift ends, the meeting is a discussion of yesterday. Yesterday's stops are worth knowing. They don't move this morning's job off the pallet. The lead who waits for the dashboard to confirm a problem the minder reported at lunch has added a round trip to bad news. The minder was the early signal. The screen was the minutes of the meeting. Treat it as minutes, and it will serve. Treat it as the signal, and you'll be late by policy.
Stops without a cause are the same kind of object. The press was down, and the tile says so, and the reason is a code that means "other", because "other" was the fastest tap on a screen the crew doesn't trust. You've learnt that you stopped. You haven't learnt whether to fix the feeder, change the stock, or stop taking that work. A cause worth having is a short human sentence, written at the time, by the person who was there. "Doubles on the lightweight, changed pallet." "Waited on plates, file was the old one." Those sentences turn a counter into a reason. Without them the dashboard invites a lecture about efficiency, and the lecture has nothing to grip.
Flattery is the other failure, and it's the one that gets dashboards defended in the face of a late lorry. Good sheets and waste in one total. Makeready called production. A reprint called a new success. The tile goes green because the definition of green was friendly. You can feel informed and be looking at a loss. Ask, once, of a job you know was bad: what did the wall say on the day. If the wall said the press had a good day, the wall is a comfort. Comfort isn't a production tool. Split the waste out. Put makeready in its own state. Put reprints against the original job. The picture will get less flattering and more like the invoice. That's the picture that can change a quote.
The next job isn't a counter
The late carton by the guillotine wasn't late because a sensor had missed it. The file had been approved late. The plates were for the previous version, and the right plates waited on a correction nobody had put on the ticket. Finishing could have creased them in the afternoon if the sheets had arrived; by the time the sheets arrived, the gluer had been set for someone else. A dashboard of press counters has no tile for "wrong plates" unless you build one and feed it. The information that would have saved the lorry was in prepress and in planning, earlier, and it didn't look like a speed or a count. It looked like a version.
Changing the next job means a different verb from displaying this one. Replan. Move the work to the other press. Hold it until the right file is plated. Ring the buyer before you print a version you'll scrap. Fix the feeder before you start the lightweight job, rather than after the waste tile has something to say. Swap the order so a heavy black doesn't ghost into a light form. Each of those is a decision, made by a person who can see the cause and is allowed to disturb the plan. The dashboard doesn't disturb the plan. At best it gives that person a reason. At worst it gives everyone else a spectator sport.
This is why a beautiful wall can sit beside an unchanged week. The meeting looks at the stops, nods, and releases the same sequence the software proposed yesterday. Nobody has the authority, or the habit, to take a job off a machine. The data has performed its role as a conversation piece. The pallet by the guillotine performs the actual role, which is the customer's. If you want the wall to matter, name the person who must act on it before the next job starts, and name the actions they're allowed to take. A lead who can only apologise for the dashboard is a commentator. Commentators don't get lorries back.
There's a smaller, meaner version of the same gap on the press itself. The console shows the minder a running waste count. Useful. The minder still needs to know whether they're allowed to stop, and who wears the cost if they do. A number climbing on a screen doesn't carry that permission. We've all watched a crew keep a press running past the point they muttered about, because stopping felt like a bigger offence than spoiling, and the dashboard would at least show a high output. Output was the wrong god. The next job, waiting, would have used the time better. Seeing the number didn't make that trade. Somebody has to have made the trade in advance: this much waste, then you stop and call.
Goods-in is where this shows up without a single press sensor involved. The right board was in the yard, still wrapped, because the delivery note didn't match the ticket and nobody had time to argue with it before the shift started. The dashboard, meanwhile, showed the press idle, then running on a substitute the warehouse had offered as "close enough". Idle looks like a problem the wall can see. The substitute looks like a recovery, and the tile goes green again. The next job, which needed the wrapped board and a different makeready, is now behind a run on the wrong stock. Nothing in the counters says the stock is wrong. A person who walks to the wrapper can say it, and can hold the press for the time it takes to check the delivery against the ticket. If that walk isn't someone's job, the wall will congratulate the recovery and the buyer will meet the substitute in the finished work. You can hand the screen a status that the press is running. You can't hand it the fact that running, this morning, was the mistake.
The same gap sits between the hall and planning when two jobs could trade places and nobody has the nerve. A short digital job is blocked on a missing aim. An offset form could move forward and free the finisher before the carrier cuts off. The tiles show both machines. They don't propose the swap, and they don't give anyone cover for making it. The planner who has been told the sequence on the screen is the plan will leave the blocked job blocking, because moving it would make the dashboard look untidy for an hour. Untidy and on time is the better run. The screen can't prefer it. A person has to, and they have to be praised for the lorry, not for the green tile they disturbed.
Make the view arrive where a decision still exists
The dashboard earns the wall under a few plain conditions. The count is tied to the right job, and you've checked it against paper often enough to believe it. Waste isn't hidden inside good work. A stop carries a cause a person wrote. The numbers show up while the next job can still be changed, not the next morning when it can only be regretted. And they show up in front of someone who can change the plan: the production lead, the planner, the minder, whoever your plant actually obeys. A screen in a corridor that managers pass on the way to something else is a poster. A screen at the planning desk, opened at the handover, with the late jobs beside the counters, is a tool.
You can hand that screen the pulses and the causes. You can't hand it the altered plan. The plan alters when a person moves a job, rings a buyer, or holds a plate. If those actions don't become easier because of what the screen showed, the screen didn't touch the run. It touched the meeting. Keep the meeting if you like the meeting. Don't tell yourself you bought a better press day. The press day gets better when the next job is a different job from the one you were about to run badly: right file, right stock, right machine, a feeder you've already decided to fix, a waste limit someone owns. Counters can help you notice. They can't sign the change. The lorry only cares about the change.