Sequence print shop orders to protect makeready

What to take away

Makeready is what sequencing can change

Makeready is the expensive part of a print job, and it is the part sequencing can change. Order the day so that setups repeat instead of reset. Five rules do most of the work.

On a single day, three jobs on one sheet run back to back, and the stock changes once. Within each stock, jobs sharing an ink run together, with the light-coverage job ahead of the heavy one. The same grouping then repeats at the folder.

Grouping buys hours that no amount of hurrying recovers. The last rule is where schedules break. A plan that assumes everything runs first time is not a plan, and the shop pays for the optimism in overtime and in dates renegotiated by phone.

Sequence only jobs with an approval

A sequence built on unapproved jobs is built on somebody else's calendar. Between order and approval sit the file, preflight, correction, the proof and the customer's own decision-making. Three of those five steps are outside your control.

So write the lead time as working days from approval, put it on the quote, and hold to it. Late approvals then move the delivery date automatically rather than becoming your emergency.

Record the approval itself with the job: who approved, what artifact they were looking at, and when. That record is also the defense if the finished job is disputed, which is why it belongs in the schedule and not only in the file.

Count back from the dock for each approval date

Most shop schedules are built forward from the order date, which is why they slip. Build each job backward instead, so the sequence you chose has a date it must meet and the customer sees which date they hold.

Work backward from the customer's needed delivery date, subtracting shipping and any delivery you make yourself. Subtract packing and hand work, the stage everyone forgets, which triples when three jobs arrive together.

Subtract finishing at the finishing department's own realistic rate, press time including makeready, and stock lead time where material is not already on the floor. What remains is the latest date proof approval can happen. Tell the customer that date.

Sequence the finishing line, not just the press

The press is visible, so it gets scheduled. Finishing is where the work stands. Grouping on the press only helps if the finishing line is grouped to match, because that is where a delivery date is won or lost.

Cutting, folding, stitching, drilling and hand assembly are frequently one machine or one skilled person each. A press clears a job in an afternoon and hands it to a folder that needs two days, and a schedule built on press hours shows none of that.

Find the constraint by looking for the queue rather than the activity. Whichever stage has work standing in front of it is the one whose hours are worth sequencing. Load that stage to a realistic level and let everything upstream be paced by it.

Release work at the rate the constraint can absorb. Everything else is a queue with worse visibility. What each machine can run, at what rate and with what changeover, comes from the presses and finishing lines themselves.

What a rush does to the sequence

A rush is not a fee. It is the purchase of capacity that was already sold. Inserting one breaks the stock and color grouping you built, so count what it consumed before you decide what to charge.

Add those up once, honestly, and you can see what the rush actually consumed. On overtime, the Department of Labor states that employees who are not exempt covered by the Act must receive overtime pay for hours worked over 40 in a workweek at not less than time and one-half their regular rates.

That count shows how many rush jobs a week the shop can absorb before the normal schedule breaks. Decide that number in advance, then see how a price follows from setup and run in the print shop price list.

Check the sequence against what happened

Keep three columns on every job: promised, actual, and where it waited longest. After a month, sort by the gap. The pattern will not be random. It will cluster on a job type, a stage, or a customer whose approvals are always late.

Fix the assumption that produced the wrong promise rather than adding a general buffer to everything, because a general buffer makes you slower on the work that was already fine.

Those job records are ordinary business records. IRS Publication 583 covers keeping business records, and they are what turns a scheduling argument into an arithmetic one. For the map of every place a job can wait, see the workflow guide.

Common questions

How much slack should a week carry?

Enough to absorb one ordinary fault without moving a promise. Shops that carry none spend the saving on overtime, and shops that carry too much lose work to competitors quoting shorter lead times.

Should I quote a range or a date?

A date, from approval, on a stated basis. A range gets remembered at its shorter end by everyone who reads it.

What do I do about a customer who always approves late?

Tell them what it costs them, in writing, in the form of the delivery date moving. Most late approvers are not being difficult; they simply do not know the clock has not started.

Is scheduling software worth it for a small shop?

Only after the sequencing rules exist on paper and are being followed. Print MIS and scheduling modules such as EFI PrintSmith, Avanti Slingshot and ShopWorks OnSite schedule what you tell them to. Software applied to an undecided process produces a very detailed picture of a schedule nobody keeps.