What Makes a Good Designer?

My hero, the baseball guru Bill James, was able to look at the same questions that many people had thought about for many years and lay bare their elements. Yes, he created all kinds of new statistics, but what really made him special is how he could bring clarity. He knew how to simplify the problem, and then skillfully describe a logical path to an answer. When he was confronted with the question of, “Who belongs in the Hall of Fame” he realized that some players were outstanding, but only for a few years. Other players had very long careers, but rarely were considered candidates for the “MVP” award in any given year. I’ll apply his conclusions to the question of “What makes a good designer?” and offer up a few more baseball analogies along the way.

Some attributes that I’ve heard used in job descriptions to describe a good designer:

Intelligent

Proven leadership qualities

Motivated / strong work ethic

Detail oriented

Team player

Flexible

Creative problem solver

Computer skills

Experience

Knowledge of construction and or / engineering practices

I get a chuckle out of some of these. Who isn’t going to describe themselves as “motivated?” And I translate “flexible” as “willing to work nights and weekends, if we ask you to.”

Rookies

I was a truss designer for about 10 years. Prior to that I had worked in the shop – mostly on the radial arm saw. Later, I spent 2½ years training designers how to use MiTek software. After all of that, I would say that there is no way to predict ahead of time whether or not a person will be a good component designer. I would even go so far as to say I don’t know if there is any way to predict whether or not a person will even like being a component designer. Still, everyone has to start somewhere, sometime, so here is my description of the ideal rookie component designer.

Thinks that a job designing components would be a really cool job

When I went from being a radial arm saw operator to a designer , I thought, “This is great!” Compared with a lot of job’s I’d had – this was a step up. Working with design software, doing important work (holding up people’s roofs!), doing something that not every Joe Schmoe could do… that was all good stuff to me. Looking it at it the other way, you don’t want someone who is simply doing this because nothing better is available - at present.

Learns the software so fast it makes your head spin

I taught people who learned quickly, and those who struggled. The basic truth is that component design software is not that difficult to learn, if you have an aptitude for it. There are going to be exceptions here – people who through dedication and effort overcome their initial struggles and become great designers. There will also be people that find learning all software is easy, and after quickly “mastering” design software they become bored and quickly look for new worlds to conquer. That been said, I would not be very patient with a rookie who struggled to learn design software.

Prior experience in the shop or in the field

Understanding the challenges of field framing is invaluable. Understanding the challenges of your own shop is equally invaluable. Having either experience gives the person built-in empathy for the customer – either the internal or the external one.

Bill James showed how star players rarely had their best seasons as rookies. It usually took ballplayers two or three years to hit their stride. Still, even in these first seasons they established that they belonged in the big leagues and rapidly improved as the years went by.

Veterans

When I worked as part of a large (15+) design department many years ago we had good designers, and not so good ones. What was the difference?

Experience

This has to be the #1 attribute of a good designer. The experienced designer should be more productive, make fewer errors, and be more valuable to others. Experience can be a tricky, though. I had years of experience before I did my first layout and learned to interpret plans. Some folks find a comfortable niche and just hang out. Those folks are long on experience, but short on performance.

Professional pride

The guys who take the job to heart, who see themselves as design professionals don’t need any further motivation to show up and do good work. They expect it of themselves.

Focus

A favorite designer I worked with leaned how to use all the hot keys within the software. Watching her work, you could see the payoff. I had never seen anyone work so quickly. She was not interested so much in optimizing each component to perfection, she was focused on getting work to the shop. In baseball, some guys hit for average, some for power, A few can do both. My friend Debbie hit for average – she stayed on task, refused to get side tracked, and got work to the shop faster than anyone else I’d seen. Guys who get too distracted experimenting with the software or adding too many details to their presentation drawing may think they are doing great work, but they could be more valuable if they stayed on task and focused on getting the work out.

Career Value

Getting back to the Hall of Fame question, Bill James realized that each ballplayer had a “Peak Value” and a “Career Value.” Peak Value was how good they were at their peak. Sandy Koufax and Sammy Sosa had very high Peak Values. Career Value is cumulative, it can’t be measured until the player’s career is over. Don Sutton and Paul Molitor are players whose Peak Value was not very high, but got into the Hall of Fame on being quality players for a very long time. The best designers are building up a high Career Value by showing up, being dedicated, helping others, and making the office a better place to work in.

Interview with Casey Carey

Many of you already know Casey, or know of him. He is a 30-year veteran of the component industry who lives in Indiana. I became acquainted with him listening to him speak on wall panel production at a BCMC Show. Later, while working for MiTek in St. Louis, I was able to visit him while he was Operations Manager for the Stock Building Supply plant in Franklin, IN. Casey is currently a consultant for RevisionLLC. I consider Casey to be one of the best critical thinkers in our industry. He graciously agreed to an interview for this newsletter. Here are some of the highlights of that interview.

Core Strategy for Production Management

GI: Is there a particular concept or strategy that you consider critical to effective production management?

CC: I think it is essential to have an accurate scheduling system with measurable benchmarking.

GI: There are a lot of different ways to do scheduling and benchmarking. How do you choose which one to use?

CC: Any method is acceptable, as long as you are constantly working to improve it. You want a create a system that is not going to cost you a lot of time and money to maintain and you need to integrate this system into your manufacturing process. You don’t want a system that is dependent on one person keeping a lot of records. In that type of system, if that person misses a day – nothing is recorded.

GI: Scheduling can be done with 3x5 cards on a board in the production office. Is this an acceptable scheduling system?

CC: Jobs have lots of different characteristics, and a name and a number doesn’t tell you much. If you take the “cards on the board” idea to the next step by making your floor truss jobs blue and your roof truss jobs yellow, well, that’s more at least a little more information. The more sophisticated you want to be, the more characteristics you will want to measure. I want to be able to use those characteristics to learn more about what we do well, and what we don’t do so well. I also like to look at the various stages a job goes through, and create benchmarks for each stage. When we do that, we have a better chance of discovering where our bottleneck is. It’s important when measuring that the reporting be accurate. Without it, you can’t tell if a variation in the data is ‘normal’ or a special case that merits investigation.

Truss production is picking, cutting, manufacturing and packaging. It took me a while to fully realize that when I had finished building the trusses, the product was not ready to ship. A number of steps, such as sheathing gables and getting the paperwork and the hangers together, are all part of the ‘packaging’ process. I also decided it was important to know, for each job, how long the trusses were sitting in my yard before they were shipped.

GI: How did you find that useful?

CC: It begins with the notion that you don’t want to build something and have it sit in the yard where you can’t get paid for it. In measuring “days sitting in the yard,” you may discover how accurate your customer’s scheduling system is. Sometimes it reveals the habits of the sales rep. Did he get it into production before it was needed just so he wouldn’t have to manage it anymore? When I was working with production builders, it was very helpful to have a complete report of what was sitting in the yard, and how long it had been sitting there. If he was hammering me with “When I am I going to get Lot 12?” I could respond with “Why don’t to also take Lot 14? You ordered that two weeks ago and it has been sitting in my yard for a week?!” Measurements like “days in yard” help you determine if you have a system problem or a people problem.

‘Just in Time’ Production

GI: Is a ‘just in time’ production process practical for truss production?

CC: Absolutely.

GI: What about the argument that you need to get a day or two ahead on the saw? In a JIT production system, if your saw goes down, aren’t you at risk of having to shut down altogether?

CC: A lot of people think production is a cutting issue. They give sawyers incentives and overtime that encourage them to get ahead, then the pieces sit in the yard for a day or two. I think you need to eliminate tying up capital like that. There’s also the hidden cost of the extra time the guy who stages material for the tables takes to find the needle in the haystack of cut pieces. The guy at the back end of the saw should know exactly where to take the pieces to, and the guy getting the pieces for the tables should know exactly where they are without hunting around.

Ideally, you have people cross trained so you can put more people where your current constraint is. And if you are measuring things and know your expectancies, you can foresee when you should only be operating one saw or when to move your sawyer to another task. Getting way ahead with the saws is an example of doing ‘a wrong thing right.’ On the other hand, implementing a lean manufacturing process and finding you are running short of pieces for the tables would, in turn, be an example of doing the right thing wrong. Having clarity about what the ‘right’ and ‘wrong’ things is tough, and an ongoing process.

To the point about managing “in case the saw goes down;” how often does that really happen? If it happens fairly often, is it a problem with the equipment or are you just not doing what you should be to maintain it? When I managed a plant, my rule was, “If I can’t get the part in three hours, I’ll buy an extra one and put it on the shelf.” I was confident the cost of those extra parts was cheap insurance compared to the cost of that piece of equipment being down for a day or more.

How much does it cost you when a piece of equipment goes down? What does equipment down time cost you annually? Grappling with these things isn’t easy. If it was, everyone would do it. On the other hand, sometimes people just need a little illumination about the right things to do - and it gets them headed in the right direction. Many times there is already someone in the plant who knows where an improvement could be made, but just doesn’t act on it; in some cases it’s because he is too invested in the current process.

One attitude that is so important within the organization is taking care of your internal customers. If you are the saw department, are you doing everything you can to make sure that your customer, the table guys, have everything they need? Do they know where to get work for the next job? Is it clearly marked? Do they know where it should go on the production line? Treating the next guy in the manufacturing process as your customer and constantly trying to improve the service you provide to him will make your overall process better and better over time.

Business Management

GI: Do you think there are any common misconceptions or blind spots in the industry right now?

CC: I think CM’s are blaming poor bottom lines on the competitiveness in the industry right now, rather than looking for opportunities to become the low cost provider. Think about:

· What was my estimate cost?

· What was my design cost?

· What was my build cost?

· Did I get the job to the site on time?

· What ‘value adds’ am I performing that I don’t have to be doing?

· What ‘value adds’ do I need to do that I’m not doing?

By looking at each of these carefully, we can all find areas where we can lower costs. Also, by really working on safety, it is possible to save thousands of dollars a year on workers comp claims. By making a study of ‘errors and omissions,’ you can eliminate many of them. It wasn’t until I looked hard at my errors, that I realized that most were coming from design. That let to steps to reduce the errors, and lower our costs.

GI: Any thoughts on the effective use of the software tools that are out there?

CC: Scheduling and inventory control software should be used to their fullest capacity.

GI: Why don’t you think people make better use of the software they have?

CC: Two reasons. First, they don’t know how to use them. Second, they are fearful that the software will tell the whole story and they will end up looking bad.

An example of the type of measurement that today’s software has the potential of providing: We figured out there was a fairly high degree of probability that certain plates sizes that were going to be used at a certain locations on our 110’ line, for example we used a lot of 5x5 plates at the peaks and many times the peaks fell in roughly the same place. We put a bunch of those plates in bins at those locations on the table, and then only picked the plates that were “special.” It saved a bunch of time on plate picking. We couldn’t have done that without the software’s help.

With inventory in particular, I’ve seen several things derail implementation. In some cases we just don’t want to do the work it takes to set it up and maintain it. In others, the purchasing people don’t really want to work that closely with manufacturing. I’ve even seen cases where it just comes down to not having everyone on board with the plan. Managing inventory is crucial, and the software can help a lot, but it has to be used in order to help.

Lineal Saws, Auto-jigging, ROI

GI: Lineal saws: are they the future, or simply a nice compliment to a component saw? What about auto-jigging?

CC: If you are talking about replacing a component saw with a lineal saw, be sure to consider the considerable changes that probably will need to be made to the layout of your facility. Lineal saws have several facets, making it hard to make a definitive statement about them. Let’s look at it this way, we want to make greater profits, right? Greater profits come through less inventory, less cost, and more throughput. Lineal saws can help in each of those areas but they need to be thoughtfully implemented with each of those factors in mind. The ROI on lineal saws and auto-jigging is highly dependent on the average quantity per setup your plant runs.

GI: How do you go about analyzing ROI?

CC: When looking at ROI, I like to boil it down to man hours saved vs. how much more (or less) work I am going to get out. I am also mindful of the ‘hidden savings’ that come into play when we can improve accuracy, quality, and safety. For example, if I cut more accurately I’ll probably be able to set up a little faster and produce a product that performs better in the field.

What’s Coming in the Future?

GI: What do you see coming in the future?

CC: I see one guy who does all the design work; trusses, panels, EWP, the whole thing.

GI: A lot of people say they don’t have enough time – we will always need more than one person working on a job to make the delivery on time.

CC: I think we are guilty of not giving our people time to learn new things. Look at BCMC – a lot of owners but very few designers.

GI: What else?

CC: I see CM’s providing more things for the customer, things like rake ladders and stud counts even when we don’t supply wall panels. And I think more plants will be doing wall panels. I also foresee software that ties together all my equipment in the plant with all my business processes.

You can contact Casey Carey at casey1212@att.net.

A New Kind of Loading

Last week I talked about how pricing will be different in the future. This week I’ll make the same claim about the loading of trusses. One day, not so very far in the future, the trusses you will be building will be using loads developed within the model (the 3-D layout) itself. And although you might think this would only be of interest to engineers and truss designers – it may mean a whole lot more than that. This brave new world of loading adds a potential new product line to our business – information. Time will tell if these ‘information products’ will generate additional revenue, or simply become an expected part of what the “truss guy” provides at “no charge.” Now is the time to be thinking about it. More about these new products later. First, let’s review where we’re at.

Loading of trusses from the Beginning of Time to the present

Trusses are designed in basically the same way as they were 25 years ago. Key elements of this ‘old world’ are:

1. Basic loads are applied to the truss (expressed in a few values) and applied uniformly

2. Each truss is analyzed and approved in isolation from all the other trusses

Thirty years ago it was a major feat of computer engineering to model a truss, evaluate the forces and stresses and specify at least an initial solution. Layout software did not exist. Engineering software needed everything about the truss typed in on every truss. Each truss was examined in isolation because, in part, that’s all the software could do. This legacy is deeply ingrained. Consider the evidence:

1. We seal one truss at a time. How enthusiastic is your engineer to review and seal the entire roof system?

2. How easy is it for a truss designer to line up the webs throughout the entire roof or floor system?

3. Say your first 10 trusses after the gable are all “T1” trusses. The first “T1” is in the wind “End Zone.” The others are in the wind “Interior Zone.” Can you analyze the T1 for both conditions? If all the trusses are carried by a girder, is the girder loaded accurately, or with the “worst case?” Isn’t the girder highly overdesigned (think uplift) if you do that?

4. Does your labor estimation formula take into account how similar (i.e. same pieces) the different trusses in a particular table production group are?

None of these four things is “easy” because the software, still looking at each truss in isolation, does not easily lend itself to these situations.

What’s wrong with this picture

Good ol’ simplified loading works. It would be nice if we could continue to live in that world, but we can’t. Beginning more than 20 years ago, the introduction of wind loading began the march towards where we are today. Picture a banner held high that reads, THE MARCH TOWARDS EVER MORE ACCURATE LOADING OF YOUR TRUSSES being carried by the building code specifiers of our world. To accurately design your trusses today you have to (for each truss, mind you) specify (manually) the Height Above Ground, Wind (answer about four questions), Snow (answer about 5), bearing material, bracing material, building code, AC loads, light storage loads, drag loads, sprinkler loads, unbalanced loads, and so on. The Jeffersonian in me particularly likes the codes that require that open sections of panels be designed for storage loads, even if no homeowner in their right mind would crawl deep into a hip system to store their blankets and old cook books.

The engineering people would have us load our trusses accurately. Well, we’ve reached the limit of what we can do looking at each truss in isolation. The newer building codes, specifically as they apply to wind and snow, make it impossible to accurately load the truss without taking into account where the truss actually is within the structure. We have arrived at a place where we need to model the structure, and have the model inform the truss design engine all the pertinent information so we don’t have to. There’s just too much of it to do it accurately.

It’s Coming

This model-driven loading is coming. By this time next year (if not sooner) it will be available within the MiTek software suite. Other software suppliers either already claim to have it, or will, shortly. I’m convinced it really will be a better world, once the kinks are worked out. Each individual component will have the correct loads applied and load paths will be traced throughout the structure. I expect that once in use and established, that building codes will begin specifying that that each component must be designed for its exact condition of use (where it is within the structure), rather than simply as “20-10-10 with 90mph wind and 80 lbs. Ground Snow.”

What’s the Benefit?

First, it should be said that this doesn’t have to have a benefit – it’s coming regardless. It is more accurate, and the last 15-20 years have shown a relentless march to more accurately loaded trusses.

But there will be benefits, and I want to list the ones I can think of here. You will likely be able to think of many others.

Benefits for designers

· No longer have to hand calc

o Floors carrying wall and roof above

o Wall girders

o Girders carrying both roof and floor loads

· The information common to all components (building code, structural materials, wind and snow rules will reside in the structure.) Change the data in one place and have it promulgate through the components in the structure.

Benefits for the Business (“Information Products”)

· Ability to easily calculate the loads for:

o Headers

o Beams

o Columns

o Foundations

· Other information useful to the builder

o Square footage of roof and floor decking

o Rough opening takeoff

o Drywall square footage

o Lineal footage for trim, fascia, gutters

OK, I’ll admit these last ones have nothing to do with loading, but the point is that once we have created the model, one of the returns on that investment is ready access to lots of information about the stuff in the model – both structural and cosmetic. Be thinking now about what information is in the model that is valuable to someone you know. Can you get that information now? If not, why not ask that it be provided? That’s what the software is there for.