Enclosure Build Cost Estimating

We are an OEM panel builder. UL508A and all that. We've been in business for over ten years and shamefully we are still using empirical guess work to estimate hours for our labor. I'm not convinced this is the most efficient way. I'm aware that many shops use a cost/point system for labor but we've been at a loss to arrive at a value that makes any sense.

We'd appreciate any feedback that the forum members could provide.

Thanks,
... rab

I worked for a facility that had a panel shop a few years ago. The shop foreman used 5 min. labor/each digital point termination and 10 min. labor/each analog point termination for estimating, plus construction labor time to assemble things before landing wires. His estimate for points included time for measuring, cutting, stripping (and heat shrinking as necessary), and terminating the two wires for the digital point and the three for the analog point. We had pretty experienced assembly techs, and they worked pretty quickly.

rytko ...

Thanks, this is valuable stuff. We work quickly too, but none of us has shown that we can quote the labor without guidelines. Parts, no problem. Sometimes we win, sometimes we lose with our current method.

... rab

Tracking

I am the VP of an OEM panel shop and have had the same problem. The difficulty with using someone else's "standard" methodology is this: Your employees work at different speeds and are at different levels savvy wise; this makes using a standard system unwise.

I suggest this: (works well for us)

On your next project, keep an extremely careful log (Excel spreadsheet for me) of exact amounts of time for each component. Then either keep the install time required on a spreadsheet and have two tables; one for components price, and one for installation cost.

Or if you have someone who can, you can do what I did, and plug these numbers into a VB (Visual Basic) program and it will do it automatically.

For an example, we build VFD systems (Variable Frequency Drives) and my program looks like this:

When the program starts, you see an empty enclosure, and to the side of it are components such as drives, line reactor, fans and filters, breakers, contactors, etc. (All the standard things you find in our cabinets). From here you would click on a component, say a drive, which would bring up an option box for you to choose voltage, horsepower, etc. it then sizes the rest of the components automatically.
So then you simply drag all the components into the enclosure to make sure you didn't forget anything, and click "Price" and there you have a total. (Our cost and a quoted price)

Of course this is only possible because I took the time to figure out exactly how long it took OUR shop to install components.

Hope this gives you some ideas....

David Groce
Vice President
www.kraftronics.com

Thanks, David ...

The only difficulty that I might have is that we haven't had a typical panel that we build for one or many customers. All of our panels have been custom and contain many different components (as do all panels but I mean components not only unique to the project but unique to us). We may have done only 1 or 2 panels ... ever of a type.

However, we are now getting requests for multiple order quotations (hence the thread) and haven't developed a scheme to handle these. Many of our customers let us T&M the builds. Of course this makes life easier but it's too much to expect of a new client that doesn't have a history with us.

I can see, however, that the special components may not be that big of an issue if I've prepared a costing basis beforehand. I can handle the VB code and/or design a spreadsheet and may yet explore this.

... rab

I was the Design Engineer in a shop that produced Control Panels for High Speed Plastic Processing (Relatively High Speed - I've since been through some Pick-n-Place System Designs that made the Plastic Systems look like they were standing still).

Anyway...

At the "Plastic" shop, almost all of the cabinets were "one-off". That is, one time design, one time build. Essentially, a vast majority of the jobs, in one way or the other, was a Custom-Build.

All of those jobs were "the same... except".

We could nail down the "REAL" basic layout. Certain control buttons applied to all, certain hardware (drives and power supplies) applied to all (sometimes).

An order would come in and a "BARE-BONES-MINIMUM" cabinet was built. I was working on the drawings at the time that the cabinet was being built. Sometimes we found we had built "BARE-BONES-MINIMUM - VERSION-A", when "BARE-BONES-MINIMUM - VERSION-B" was really needed.

If Version-A was built, and then it was determined that Version-B was needed, sometimes parts had to be removed/relocated/replaced.

We found that sittin' around having a cup of coffee and a smoke and/or a Danish, and discussing the up-coming project for an hour or two saved us a lot of grief.

I tried to have a heads-up on all of the design issues, but, having to do "right-now" field-service calls and writing documentation for previous systems at the same time, sometimes I was a little out of touch with the current project (over-load?).

I had some damned good people working for me. They saw clearly what I sometimes saw as a fog. Their input saved our a$$'s plenty of times.

Having that input, we could nail down the components that were a given. They fell into a particular real-estate scheme. That much could certianly be done on the floor until I was able to handle the remaining design issues.

Eventually, after much data gathering, the real issues made their way up to sales and they began to get a grasp on the realities of custom-design.

Now, as far as costing the panel effort, over the course of trying to make proper decisions regarding "basic" panels, the hardware nature of the system was more clearly defined in terms of the functional nature of the system.

That is, just as "Form Fits Function", we were able to develop a general guide-line in terms of the required function. That became the primary costing factor. All of the incidentals, wire landings, et cetera, were "ball-parked" in so as to, at least, break even.

A system that required 800-wire-landings was significantly, functionally, different from a system that only needed 100-wire-landings. The fact that there were 800-landings was actually a function of the required function!

The key to the game was, Costing by Required Function. And, of course, the REAL cost can only be obtained through experience.

We found that the idea of maintaining a "cost by landing" gave us more grief than simply "ball-parking" the landing-cost by function.

Another primary issue, as David indicated, is that you have to have people (or machines) that will perform as required (at least, reasonably so - don't get me wrong! I'm a Democrat, maybe, more of a Socialist, I DO NOT BELIEVE IN SWEAT-SHOPS!!! Geeezzz, Is that Tangent-Bait, or what??? That definately bears a thread of it's own, if any of you have the HAIR!!!).

At any rate, the name of the game is to make "wire-landings" insignificant in terms of the cost! Go for the "Required Function" and the "ball-park" figure to handle the incidentals!

Tangent

That definately bears a thread of it's own, if any of you have the HAIR!!!

Click to expand...

I really didn't think you wanted to mention Da Bears this season.
Ouch!

How bout the St. Louis Comeback. Are they for real?

;-)

The Bears have unveiled a new logo.

It's a rebuilding season.

I design panels for plc's, drives etc.
How I present what I want to a panel builder bidder is extremely important in helping get good bids where everyone wins (I get what I want and the builder makes money).
I have standardized on three basic parts:

1. An accurate layout diagram with complete bill of materials
2. An accurate and easy to follow interconnect diagram (using schematics to generate interconnect)
3. Give guages for all wiring.

To me, having accurate and complete information is the key, finding a way to get this maybe your answer.

Hope this helps

George,
Just in case you might have misconstrued my intent...
That's not exactly the potential-tangent-thread I was alluding to.

But I do have to admit, that new kid is amazing!

Rick,
OUCH, OUCH, OUCH!! DAMN IT, I KNOW IT'S TRUE, BUT OUCH, OUCH, OUCH!!! "DA CRIPS"


Glav,
We've been "re-building" since 1985! I want to see another Ditka at the helm, or better yet, another George Halas!

Yeah, Lefty, and you could have what's-his-name (the guy with the teeth) back!

But then... they ain't the touchy-feely kinda coaches that are acceptable these days, are they?

I just can't wait 'til Mister Rogers (he's the cause of all this, don't cha know?) becomes a boot-camp Drill Instructor for the Corp!

It's a happenin' thing!

To quote one of the regulars.... "...Sigh..."

Thanks to all for their input as well as an amusing diversion to my old hometown favorite 'Da Bears'.

My intent was to get ideas on how others get around the intangible elements of quoting an ICP (industrial control panel) ... labor. I have an excellent staff to work with and imagine that on a one to one basis, we can compete timewise with anyone but to determine on a new project what that time might be is still extreme guestimating.

BOMs and schematics are certainly valuable and cut down the research requirements but can be determined from the control concept (napkin sketches and the like). It's this labor thing that has me. I'm a formula kinda guy so the per point method intrigues me.

Any one else with some formula methods?

... rab

Estimating

Believe it or not a PROMINENT drive manufacturer with which we do business follows very simplistic path.

They take the cost of ALL components in a panel a divide that by .9 for installed cost. Simple as that.

(That makes me nervous personally, but they're very successful with it.)

David Groce
Vice President
www.kraftronics.com

Actually that's how I've always done it. Take the cost of the components and double it.

Estimating

Double it? There's a huge difference between /.9 and (x2) you've been succesful with that? What components do you typically use?