Crates come in all sizes.

This one is from our “extra bedroom” series.

Looks like we might need that handy expand- a-truck tool…

Just another day at the office.

Our sympathies, guys.

Betsy Dorfman

In the not too distant past our customer service folks, with degrees in things like medieval literature and art history, were forced to rely on various musty charts, formulas,  incantations and dart boards in order to produce crating cost projections. Sometimes reality obliged and we came close; sometimes not. On average, we weren’t. But with the arrival of  crate wizard Chris Barber, all we have to do is send an email and wait for his reply. Here’s more from Chris on the nuts and bolts of his creation:

Depending on how comfortable you are with creating and managing a partially automated system, a custom estimate and cut-list program can be a ridiculous time saver for your crating department. My “crating engine” uses mostly simple math functions in a simple database application. With it, I can estimate the cost and dimensions of a crate and have a formatted cut list ready to print for the craters in as little as fifteen seconds. Unusual crating circumstances only require a couple minutes of data entry before the results can be sent to customer service representatives or printed for execution. The same artwork specs and basic packing approach are automatically forwarded into several crate shell styles simultaneously, from slat crates to our highest-level travel crates. Every square inch of building material is automatically added up and priced, both for estimates and for the actual price of the built crate.

But whether you have your own crating program, or whether you do all of your math with pencil and paper, the big unknown for crating estimates is labor. Any given builder will have good days and bad days. Averaging their past performance won’t always give a perfect estimate, but it will take their history into account and mitigate guesswork based on misleading examples. Naturally, the more examples of past performance you record, the more likely you are to approach a good reliable mean.

The other sticking point in estimating labor is the duration/volume ratio. For obvious reasons, this ratio is not a straight line, but a curve. The smaller the cubic footage of any style of crate, the more minutes it will take to build per cubic foot. Likewise, the same curve levels off to nearly flat in the upper size range. I’ve plotted these curves for my lead crater so that I can make a reliable prediction of his performance on any style of crate, regardless of the size job. Even if you do everything else in your head, an accurate time curve is an elegant alternative to guesswork. Of course, this isn’t limited to crating. It can be applied to any production task with a similarly predictable set of actions. Here’s how to make your own:

Step 1. The first thing you will need is the raw data. Start recording exactly how long it takes you or your staff to build crates. Start a separate log for each crater, and each style of crate that crater produces. Every log should include a series for minutes and a series for cubic feet. Then make a third series, dividing minutes by cubic feet. I put these series in columns; so if cell A3 = minutes, and cell B3 = cubic feet, cell C3 = A3/B3. You will only use the second and third columns in the next step – cubic feet & minutes/cubic foot. Here’s an example log for “B-crates” with two hypothetical craters, one a faster builder than the other:

Soon you should have enough data in those series to get reasonable estimates. The data collection is an ongoing process, however, and your logs should be updated regularly. Older numbers could be dropped eventually to account for your crater’s growing experience and speed, but the aim is to collect as much information on each builder as possible. This is not to spy on your crew. It is to accurately predict the time it will likely take this person or that to build the next crate.

There are two ways you can process your database into functional labor estimate curves. First I’ll show the quick way, and then I’ll explain what these numbers mean by showing the chart method.

Step 2a. Find the “power trendline” of each crating log you have made, and multiply it by the estimated cubic feet. I’ll explain what the power trendline is in some depth below, but for now you can just treat it like a magic spell. If you aren’t a math geek and don’t care how, why or whether this really works, you can stop reading at the end of this step.

The fastest and most efficient way to process a given crater’s average curve on a given style of crate can be done in five math functions, and will fit on a spreadsheet the size of a postage stamp.

cell A1: =[length]*[width]*[height]*1/1728 [estimated cubic feet]

cell A2: =EXP(INDEX(LINEST(LN(y),LN(x),,),1,2)) [coefficient A]

cell A3: =INDEX(LINEST(LN(y),LN(x),,),1) [coefficient b]

cell A4: =A*(x^(b)) [trendline equation]

cell A5: =[cell 1]*[cell 4]*1/60 [labor estimate]

A1) The first cell should simply display the cubic footage of the crate being estimated. The least fussy way is to link this function to three blank cells somewhere else where you enter the crate’s L, W, & H. Those same three blank cells can be linked to every curve you make (since you need a separate curve for each crater on each style of crate).

A2) The second cell should return the value of A to be used in the equation in cell 4. This cell should contain the exact function shown, but in place of x, link to the whole cubic feet series in your crater’s log (B3:B14, to use the slower crater shown above as an example). Likewise, y must be linked to the whole series of data in the minutes/cubic foot column of your crater’s log (In this example; C3:C14).

A3) The third cell should return the value of b for the equation in cell 4. Treat series variables x & y the same way here as you did in cell 2.

A4) The forth cell should contain the function shown, but replacing x, A, b with the results of cells 1-3 respectively. Caution: in this equation, x refers only to the cubic footage of the crate being estimated. It is not the same variable as in cells 2 & 3.

A5) The fifth cell is the product of the values returned in cell 1 and cell 4, then divided by 60.

You can use these five steps to bypass the charting step described below and get your trendline equations straight from your database. But the chart actually shows what these numbers mean, and I prefer to see graphic representations of the curves anyway.

Step 2b. If the step described above seems too cryptic, the numbers involved can be more readily understood by graphing them. The program I use allows me to insert a visual chart into my spreadsheet, define the x & y parameters and link them to the two relevant series of data. This is pretty basic, and I’m sure that it’s a universal feature in spreadsheet applications. The type of graph you want is an x-y scatter chart. Your chart’s values are simply: x = cubic feet, and y = minutes/cubic foot. Once your graph is linked to those two series, you will see points plotted in the field – each point representing the crater’s performance on a specific crate.

The more information you have (and the more consistent your crater is), the more it should suggest the hint of a curve starting in the top left corner and ending in the bottom right. Now you can give the graph a trendline. The trendline extrapolates an average curve from your unwieldy cloud of points, in a visible line. You may need to choose from several types of trendline. I prefer what my application calls the “power” type, which appears to produce the most realistic curve, leveling off dramatically as it approaches zero on each axis. The “exponential” and “logarithmic” types both trace the trendline right off the chart at each end, and there’s no way a large crate will ever take negative minutes to build. Nor will a small crate ever have negative dimensions. The “linear” type overrides the curve that I believe is there. The “moving average” type defeats our purpose entirely. The “polynomial” type creates a dip in the middle ground that doesn’t make sense to me. Even if I wanted to address the handling logistics of larger crates, this potential issue is completely unrelated to the polynomial equation.

As you can see above, there is less data from the faster builder, and the blue curve is barely visible. This makes the blue trendline less reliable in the extreme size ranges; particularly the smaller sizes. This problem can be addressed quickly by giving that crater a very small crate to build and a very large one. Getting just a few points plotted past the margins of that crating history will give the blue trendline a wider range of accurate predictions.

Step 2c. Once you have your trendline plotted, tell your graph to show the trendline’s equation (which is hidden by default). Each trendline is described by a math equation reflecting the moving average of your plotted data. The power trendline equation should look like this:

y = Ax^b

The values of x and y are still cubic feet & minutes per cubic foot respectively, as the chart suggests. The coefficients “A” and “b” come directly from the trendline, which in turn is a biased average of the data your chart illustrates.

Step 2d. Now here’s the nice part: Your trendline equation can be recuperated back into the spreadsheet for the purpose of estimating labor. Once you estimate the cubic footage of your prospective crate, you can simply multiply it by the trendline to get the most accurate possible labor estimate for any given crater. The spreadsheet function for this looks a little tricky, but here it is using the same variables, A & b, as my example of the trendline equation above:

=A*(x^(b))

So if your trendline shows the equation: y = 35.956x^-0.789

…the spreadsheet cell representing it should say: =35.956*(x^(-0.789)).

If your trendline shows the equation: y = 5.5678x^-0.2912

…the spreadsheet cell representing it should say: =5.5678*(x^(-0.2912)).

Note that to make either of these examples functional, x must refer to the cell that displays the crate’s estimated cubic feet. The current value of x must be folded into the trendline equation before it can return a relative unit of duration/volume adjusted by the crate’s size. While the trendline equation merely displays the coefficients A & b, the spreadsheet cell as typed above will return the actual value of y — as long as x points to the cell displaying the current value of x and the function begins with the equal sign. Once you have a spreadsheet cell representing the trendline linked to the variable cubic footage cell, all you need do is multiply the two cells. Keep in mind that this will result in minutes; so if you prefer estimated hours, just divide the result by 60.

So to mentally separate this step from the raw database illustrated above, let’s skip over (arbitrarily) to column H on our example spreadsheet.

The blue and orange numbers in this screenshot represent the faster and slower craters, like in the curve chart. The top number in each set is the cubic footage of the crate currently being estimated. This cell changes with every estimate, as it is the product of the crate’s length, width & height, divided by 1728 to convert from inches to feet. Let’s say for the sake of argument that the cell displaying orange cubic footage is in position H4 on the spreadsheet. The next cell down, H5, is the trendline equation for that crater, with the current cubic footage plugged into it. So in place of “x” in =A*(x^(b)), the function says H4. And in place of “A” and “b”, the function shows the actual trendline coefficients. In this case what I actually typed into cell H5 is: =70.254*(H4^(-0.656)). Refer to the orange trendline on the chart to see how I got A and b. This is a functional version of the trendline equation, responding automatically to the cubic footage displayed above it. If the cubic footage dropped, the result displayed in cell H5 would rise appropriately for the crater in question. The next cell down, H6, is the product of the first two cells, divided by 60 to convert from minutes to hours. This is the estimated hours it will likely take that crater to build that style of crate at that particular size.

Step 3. Update and fine-tune your logs. Some spikes may occur that throw the whole curve out of whack. They are usually in the negative direction – like when a crater made a big mistake and spent a lot of extra time correcting it. I toss the worst spikes. I would rather take the hit when random problems happen than let them affect every estimate. Such large spikes are very rare, and I’ve only eliminated about four crates from my whole database for that reason.

Packing estimates: Of course, packing a crate involves many more variables than building it, so you should keep building time and packing time separate in your database, charts and equations. I don’t even use packing curves myself. I use a flat time for each type of flatwork, sometimes adjusted a little for size, and estimate all dimensional items in my head.

There are many different ways you can approach the problem of labor in estimates, depending on how tight you want your estimates to be. Plotting curves is admittedly a bit anal, but quite easy to set up. And it only improves over time as you add more information.

-Chris Barber

Carole Choucair Oueijan, Layaleena, 48 x 72, smalti, 24 karat gold smalti, granite, marble, onyx, crystallino, mother of pearl, fresh-water pearl, hematite, coral, jade, quartz

I always crate artworks from the inside-out; at least in my bean, in the design stage. But the actual building can vary. Sometimes it can proceed in any order, and sometimes the crate must be built before the art is approached. It depends on whether the artwork is packaged in soft materials separate from the crate, or whether it must be built directly into the crate with a cushioned wood structure. When it’s the former, I occasionally prefer to pack the art before the crate is started. This is hardly necessary, but it can save a little desk time when dealing with a number of irregular shapes that aren’t so irregular that they require much planning ahead.

This was one of those jobs that fell into that little gray area. It just made more sense to figure out how large the package would be by packing it. The piece was composed of twenty-odd irregular sections of mosaic of variable thickness. It would happily ride flat in a stack of foam-welled trays. With such a simple packing approach, it was more efficient to sort the elements by relative size and shape in “real time,” as it was being loaded onto trays. I started with a rough guideline of 36″ x 24″ trays, and from that starting point my crater found that he could fit all elements onto 13 trays at 32″ x 24″. I’m starting to make it sound more complicated than it was. Before I knew it, the trays were packed and I had a nice boxy package to measure for the crate.

Our thanks to Carole Choucair Oueijan for her permission to include images of her artwork. Layaleena, an Arabic/Lebanese word for “Splendor Nights”, is a commission piece installed in a home in Greece. In this scene the goal was to reflect the magnificence of the Lebanese nights and lifestyle of the past. Layaleena is made out of 21 pieces and took 10 months to complete.

-Chris

Today we crated an antique object so dripping with inherent vice that I wanted to just give it a good shake and ship whatever remained in hand. It was a lacy assemblage of stressed wood with flaking paint and gold leaf, forming some kind of elaborate staff.

I thought that only wizards had these, and I must say; any wizard would be proud to lean ponderously on such a grand walking stick. I was disappointed to be informed that it was not, in all likelihood, the former property of any such character; not the least because it debunked my theory that it was the gradual (and sad) fading of magic that was causing the many unsecured parts to slowly lose their unlikely relative positions and settle into a more natural relationship with gravity.

I do admire a clever bit of woodworking, and in all fairness, it isn’t the staff’s fault that it was never meant to lie prone like a saucy odalisque; much less freighted that way. But long story short, this thing was screaming to be shipped upright – not only for the fragility of its overall construction and distribution of weight, but also for the specific joins used in assembling some of the wooden nuggets.

That some of the said nuggets were quite loose is an understatement. They were ready to leap away from the piece like rats riding the explosion of a sinking ship’s boiler room. But for a variety of reasons, shipping upright was no more an option than not shipping it at all; the latter being a suggestion I slipped anonymously under the general manager’s door this morning before tip-toeing away.

When that plan didn’t work, I was forced to ship the piece in a horizontal attitude. The point of all this is that sometimes you have to just do what you can to make sure the thing is as safe as possible within the tight budget. So here’s what we did:

A cavity pack was pointless, as we would have had to float so many little parts and still catch the more solid areas at all of the correct angels in a Swiss cheese version of (a negative image of) the piece. So vertical guillotines it would be. The staff’s lateral, vertical and axial movement was eliminated by guillotines along the shaft.

One more guillotine on the crown’s widest point took much of the crown’s weight, and also eliminated any possibility of spinning in place thanks to the hexagonal tier.

Tall, tapered bumpers reached up from below to take the weight of each successive tier in the crown.

Belts and straps of Tyvek held the loosest pieces in place while holding the tiers down against their bumpers.

Finally, a safety bumper of softer foam was placed ½” away from the tip of the staff, just in case several other axial stoppers failed.

I still wouldn’t drop this crate from the top of a tall thing, but I can strap it to  a unicorn and wave goodbye to it with confidence.

-Chris

As crating manager, I sometimes get a self-satisfied chortle out of packages sent to me from various sources, private and professional. With the Tour de France 2009 in mind, this one was more amusing than most. And also more annoying.

The front wheel of my bicycle was recently crushed in a collision with a car on my way to work. It being a lightweight road bike c. 1986, I figured I’d maintain the vintage and save money in the same stroke by getting another set of French mid-80s wheels. I love eBay.

I guess it’s good that Mavic wheels are known to be pretty much bomb-proof, because the package in which they arrived was barely fit for a local delivery. The front wheel (the one I need right away) has damaged spokes, but they can be replaced. I hope that’s the only reason why the rim looks almost as bad as the one that got run over.

The thing is, these wheels are really strong; so what went wrong here? Let’s take a look.

1. Though made for shipping wheels, the box was recycled from an earlier use. There’s a different brand printed on the outside, and what passed for interior packing had clearly lost its shape prior to this use.

2. Even if new, this type of packaging is designed to be supported by other significant factors; like bundling them in large numbers on a shipping pallet. It was clearly not designed for overseas travel on its own.

3. Due in part to the used packaging, the three parts being shipped were not adequately secured inside the box. A small bag holding the steel skewers was tossed in loose to jump around inside the box, and the hub of each wheel was jammed into the spokes of the other.

As a result, the box got crushed. The cogs of the rear wheel arrived poking a 5″ diameter hole through the box, and three spokes had somehow snapped off of the front rim.

As a bonus, the shipper decided to use stamps to send the box to California. From France. I hate eBay.

So how would I have packed this, assuming that a crate was not an option?

…Let me stress that the following solution is not something that we would do here at Fine Art Shipping. Unless the client was renting a dedicated shipping container to be loaded by us, we would insist on a full wood crate for international shipping. But let’s just say that I wanted to ship the wheels back on my own dime, while trying to mitigate further damage. Due to their odd size, I would start with a custom box, built from a couple sheets of double-wall cardboard.

1. I would cut two panels of 3/8″ plywood to reinforce the interiors of the two large walls of the box.

2. Then I would surround the wheels with 2″ bumpers of Ethafoam. I would stack another 1″ of foam on these bumpers and slot them for the wheel rims. This would keep the wheels separated and secure in their cushioning.

3. Finally I would bag and secure the third part – the skewers (wheel axles) – well away from the wheels. They could be embedded in the bottom foam bumper, or the bag could be screwed to the plywood sides.

C‘est la vie.

Hey, lookit these pretty stamps.

-Chris

Every once in a while, a crate will pass through our warehouse with those air-filled plastic feet, each in the shape of a torus with one flat side, such as Skid-Mates by Hardigg Cases. They come in several densities to support different loads, indicated by a color code. These “donuts” are occasionally used in lieu of skids in the art shipping industry (more often in the general shipping industry), presumably to add that extra bit of cushioning between the crate and the floor, or to create forklift access for loads shipped without pallets or custom skids.

For each plastic-footed crate that comes and goes without incident, one or more have shown up with a crushed or missing foot; sometimes more than one. An incomplete set of feet obviously makes for an unstable crate. Those aren’t much fun when they are tall, thin and six hundred pounds or more. This can be especially serious for anything that shouldn’t be tilted, much less rocked back and forth throughout its transit. For this reason, I like to keep a few different types of donuts around the crate shop. When a crate passes through with such a problem, I simply slap a new one on wherever it is needed.

That said, I am of the opinion that plastic donuts have no place in custom art crating. By “custom” I mean crates that you design and build to unique specifications; as opposed to prefab crates and pallets for, say, a beer bottling plant. Any crate that doesn’t pack an artwork sufficiently on the inside is in trouble before it’s built, and needs to go back to the drawing board. Why not design an adequate crate and then add a little extra cushion to the outside? Because that can easily lead to taking the donuts into consideration when designing the interior packing in a crunch. Such a thing might never occur to you or me, but somebody out there is doing it right now because they are on a budget and in a hurry. If your custom art crate needs donuts under it, you’re doing it wrong. There are better ways to give your shipment multiple and separate levels of cushioning.

But donuts are not merely unnecessary on a good art crate. There is a reason or two not to use them. As I mentioned before, they come with different load strengths so that you can take the crate’s weight into account when balancing the sturdiness needed with a bit of cushioning. But what if your crate with well-balanced donuts is dropped on one edge or corner, temporarily putting all of the weight on fewer donuts? Or what if another company’s crate is stacked on top of yours? Or two more? It would seem that the feet of higher densities would be the safe bet in terms of avoiding donut failure, but that would often as not eliminate most of the cushioning for which you have chosen to use them in the first place. However unlikely the scenarios above might sound, the fact remains that I see at least one of a set crushed more often than not.

Even when none of the donuts are damaged and they are doing their intended job, the exterior cushioning they provide can cause a tall and heavy crate to feel unsteady before any of the feet leave the floor. If a transport is not properly loaded with this in mind, it leaves the crate free for all kinds of motion during transit. Hollow feet designed for specific load ranges can always be exposed to unexpected and adverse conditions during shipping.

Finally there is the potential for damage of a more violent kind. In contests with forklifts, donuts always lose. Wooden skids can also be knocked off by wayward forks, but only if the crate is too heavy to be pushed away by the forks (or it is pinned in place by something like a wall), and if the skid is also broadsided by the shock. But in most cases wood skids are placed perpendicular to logical forklift access; giving a full-length beam of solid wood nowhere to go. So the shearing of a properly assembled wood skid is relatively uncommon. Custom skids also look better, as they can be designed as a solid, seamless part of the crate rather than just being pinned on as an afterthought.

In my opinion, donuts go with coffee – not fine art shipping.

We get requests on a regular basis to send boxed or otherwise “softpacked” artworks via package services such as UPS or Fed Ex, or by freight, to various destinations.

Our answer in 99% of the cases is, no. We don’t ship softpacked art by freight service, period. Or, (with apologies to O.J.) if it’s freight it needs a crate! For starters, the package services don’t differ from freight in terms of handling; they are freight.

Freight handling is rough and tumble, or can be. We get crates in regularly that have footprints on them from being “stood” on, sometimes a whole line of footprints showing the crate was walked on most likely to reach other objects in a truck. If that container had been a softpack or box — curtains.

Most freight shipments are handled multiple times enroute, are fork lifted to and fro, and may get conveyor belted as well. The extra foam padding used within a properly built crate gives far better “shock” protection than a standard softpack. If you have ever seen the result of a cowboy driven forklift tine striking a cardboard box, it isn’t pretty. With a severe blow even a crate might be breached, but the chances are better for a positive outcome.

In addition, freight shipments are generally not “tied in” within the transport vehicle or aircraft. Which means they can shift, turn over, or other items can lean against them or fall on them. Thus, the more rigid the container the better. 

Moreover, crates can be constructed using various tricks of the trade along the top or bottom which make them suitable to ride only one way i.e. in a given orientation as to LWH. While this is not foolproof, and we never rely on freight carriers to ship art that might be damaged if turned from the vertical, it is more possible to secure the orientation of larger crates than boxes or cartons. Any small box or package shipped via freight is guaranteed to be turned this way and that, no matter what orientation marks appear on the exterior.

There are exceptions to the  freight=crate rule. Several companies sell prebuilt shipping containers or “strong boxes” composed of heavy corrugated cardboard, seriously foam lined, and sometimes incorporating additional puncture proof linings. These offer protection approaching that of a plywood crate, but are more crushable, still. And there is the occasional very sturdy artwork that can survive double boxing in a foam lined box.

Your local pickup guy may hand carry your box out to his truck, and may lovingly place it in the proper orientation, but it’s not going to be handled that way for the rest of the journey, chances are. We see packages every day which are handed to us upside down, or where “fragile” markings are ignored with parcels tossed casually into trucks despite cautionary labeling. A crate offers superior protection outside and in and gives what can be an art saving margin for error. 

If you stick with our rule of thumb you have a far better chance of your shipment arriving in good condition. 

Betsy Dorfman

Here’s a quick update to our recent post about crating the longest urinal west of the Mississippi. Terence Koh’s sculpture arrived safe and sound, and now hangs serenely on the wall of the Mary Boone Gallery.

Credit is also due to the NYC art handler, Winchester Fine Arts, who received the crate and took care of the destination services.  Shipping an artwork of this size, weight and fragility successfully is always a team effort. 

*Photos by ArtObserved

And here is a link to more photos and review of the show. Not to mention the after party! http://artobserved.com/ao-on-site-koons-kelley-koh-curated-by-javier-peres-at-mary-boone-gallery-in-chelsea-saturday-march-4th-show-runs-through-may-16-2009/

Chris Barber