Estimating sheet counts for a quote

Somebody asks what a kitchen will cost. You need a sheet count, and you need it before you have drawn anything properly, because they are standing in front of you.

The usual method is to work out the area of the parts, divide by the area of a sheet, and add a bit. That method is wrong in a specific and predictable way, and this is how wrong.

Why area divided by area fails

The kitchen job has 28.2 square metres of parts. A sheet is 2.977 square metres. Divide and you get 9.5, so you order ten sheets.

The job takes 11. You are two sheets short before you start, and you find out on the second morning.

The division is not wrong, it is just answering a different question. It tells you the absolute floor, the number you would need if board were a liquid you could pour. Every real constraint pushes you above it: the blade, the trim, the fact that parts are rectangles and sheets are rectangles and the two do not divide evenly.

What yield you actually get

Yield is part area divided by sheet area bought. Here is what the six jobs on this site actually came out at, all of them layouts we ran rather than figures we hoped for:

Between 65.6 and 95.3 per cent. Anybody who tells you to use one figure for everything is telling you to be wrong on half your quotes.

The pattern in the spread

It is not random, and once you see it you can estimate a new job in your head:

• Lengths beat sheets. A one dimensional problem has far fewer places to hide waste. The flooring job came out at 95.3 per cent and that is normal for linear work.
• Repetition beats variety. Eighteen identical carcasses tile a sheet, and the kitchen job reaches 86.1 per cent because of it. One sideboard tiles nothing.
• Big clean rectangles do well. The eight long hull pieces on the marine ply build pack to 82.4 per cent, which surprises people who expect boats to be wasteful.
• Small parts do better than they feel. The school job, 216 pieces none of them larger than 400 mm, still reached 80.9 per cent. Small parts fill the corners that large parts leave behind.
• Small and shaped is the awkward case. The acrylic letters sit at 70.1 per cent, because every letter is measured as the box it sits in and a letter is not a box.
• Long parts on a short sheet is the expensive case. An 1800 mm top on a 2440 mm sheet leaves a 616 mm band beside it, which is most of why the sideboard sits at 65.6 per cent and comes last.

Two of these numbers need a warning. The marine ply and acrylic figures count the rectangle each piece is cut from rather than the piece, so real material use on both jobs is worse than the chart says. The sheet count is still the sheet count, which is what you are buying.

Numbers to quote with

If you have to guess before you have a parts list, these are defensible starting points:

Then round up to a whole sheet, and add one more. Not because the arithmetic needs it, but because a sheet held back is a Friday afternoon you do not spend at the merchant.

The better answer

Run the layout. It takes under a second once the parts are typed, and a parts list is something you have to build anyway before you can cut.

The kitchen firm in the first example now types the parts list at quoting time rather than at cutting time. Same typing, done earlier, and the sheet count on the quote is the sheet count they actually buy. That is the whole trick, and it is not a software feature, it is just doing the work in a different order.

What to tell the customer

Quote the sheets, not the square metres. Board is bought in sheets, a part sheet costs a whole sheet, and a customer who is told 11 sheets understands the number in a way that 28.2 square metres does not.

If the job is close to a sheet boundary, say so, and say what would move it. On the sideboard, a top at 1700 instead of 1800 would have changed the layout. Customers are usually more flexible about a dimension than about a price, if you ask before you quote rather than after.