Resilient Modeling Matures – Interview with Dick Gebhard

Dick Gebhard is a guy I met while working for Solid Edge. He has a lot of energy for a guy who’s trying to not retire. He’s the guy who came up with the idea of Resilient Modeling Strategy. RMS is a method for solid modeling that gives you editable models that you can really work with, and don’t explode when you make changes. When combined with Synchronous Technology (an extension of direct edit tools), the whole RMS thing makes the most sense. When used with history-based tools, it starts to seem like the only thing that makes sense.

I’ve written about RMS before on this blog, here and here, and a good amount when I was writing for the Solid Edge blog. Mr. Gebhard has a site devoted to the topic here. It’s like the recommendations you get about eating right and exercise – the advice is clearly correct, and if you follow it, your life will be so much easier, but rarely do people really do it. I suppose we like our models the way we like them, and the parent/child relationships we love to complain about really just make us feel better. I don’t know if that’s the case or not, but I do know getting people to follow correct advice is difficult. RMS is really some of the best advice I’ve come across, especially for largely undisciplined history-based users. History-free users already understand why this sort of thing is important, but we seem to value our freedom more than the integrity of our 3D product data. If people are reluctant to see the value in something like Synchronous Technology, it’s easy to see that they also might find Resilient Modeling too much of a good idea.

RMS is really a system of modeling that uses the skeleton method extensively. You put all of the design intent in the skeleton, and then each part refers to that. You avoid the twisted links between parts and assemblies by driving. Synchronous Technology users have a headstart over history-based users because Synch, as we call it, enables you to make changes relative to anything, and there is no on-going penalty like you pay with SW in-context relations. The parent/child relations are the ones that really make modeling changes miserable, and that’s exactly what Synch enables you to avoid. Synch also helps you because the intelligence is in the software, not in the model, so the model doesn’t have the burden of all of the extra information that has to be constantly re-evaluated, you just make the changes you need when you need them. Plus, because of the intelligence being in the software rather than in the model, and the fact that RMS highly values neutral models, Synch can work with neutral geometry as easily as native. It’s just a huge advantage. Still, RMS is CAD neutral, and works with systems like SolidWorks, even if you have to apply an additional layer of rules to get SolidWorks to work properly.

I’ve said this before, and believe it to be true. SolidWorks gained popularity by making things easy. Some of that made sense, like moving to a PC platform instead of Unix. But much of it didn’t make sense, like essentially throwing out the rule book that Pro/E users had spent a decade writing. The result is that SW models are largely the product of an undisciplined approach, and you really don’t know what you are getting yourself into if you pick up some random model from some random user. The model might have a well thought out design intent, with controls that are obvious, or it might be a hippie love-in with no structure, where the only goal was to get it to a point where it is shaped correctly, with no concern about how it reacts to change. This is the kind of thing RMS fixes. It lays down a structure so that anyone can pick up any model and make changes to it, even in the potential chaos of SolidWorks “design as you model” typical workflow. The problem is summed up in a nice CAD-neutral way in one of Gebhard’s videos. The proposed workflow is laid out in another video on his site.

RMS is a part modeling system, but it also helps you structure your assemblies intelligently. Gebhard likes to speak of his site and even the whole system as just common sense best practice rules. He has said that the method is really just a shortcut to cut years off of a users time to develop power user skills.

Also, he has given some time to thinking about drawings and document management. Gebhard sees RMS as a the first step in MBD (model based definition)

I talked with Dick a few months back, and I’m just now getting around to writing this up. My comments are in green below, an Dick’s are in black.

====================================

Good morning sir, how are you doing?

I have managed to invest four years of my life trying to change the solids modeling world with RMS. I thought it was going to be easy, just hit them between the eyes with reality and everyone “sees the light”. Wow, how naive can you be?

Yeah, tell me about it. You offer something that’s obviously superior to help people, and for free, and people just keep beating their heads into a wall. I don’t get it. So what have you been up to?

My last interview with you was about 2 years ago when you were working for Solid Edge, since then I’ve come to realize some basic truths about solid modeling and you’re the perfect person to test them on, scary as this sounds.

It all revolves around a word I seldom pronounce right; hierarchy. Hierarchy is important because it establishes precedence, master-slave, parent-child or whatever you want to call it and this is the best way to capture design intent. When you eliminate hierarchy, your ability to capture design intent suffers.

Yeah, I can see where looking at it like that is helpful, although we usually think of parent/child relations in CAD as an obstacle.

I believe in the concept of the feature tree when everyone else is trying to eliminate it. It’s like shooting the messenger.

Solids modeling software vendors create their software to allow the designer to create any feature anywhere in the feature tree and link to any previous feature. This means that the designer is 100% responsible for the sequence and linking of the features, in other words parent/child relationships. Since feature sequence and linking completely determine a model’s stability, an unstable model is 100% the fault of the designer

Horizontal Modeling eliminated links in models and Direct Editing eliminates feature sequence, but both strategies are about removing hierarchy which sacrifices design intent. What’s sad is the problem is fixable, but most designers don’t care and their management accepts this.

I agree, most of the designers say they don’t have time to edit designs, but that’s a lame rationalization. Do you have many converts?

I’m making progress worldwide but it’s hard to measure because nobody cares how you create your models. Consider this, your models aren’t checked, A stress test isn’t required and you don’t have a “Best Practices Manual”. So, the converts just do it and don’t share their intentions with anyone. If it works, you can share with your coworkers.

I am also partnering with Action Engineering (Denver, Co) a leader in supporting companies that are implementing Model Based Definition (MBD). When you implement model based definition, you add value to the model (digital twin) by embedding the drawing data (PMI). If a model is recreated because of editing issues, the digital thread is broken.

I see you are calling your web site a Best Practices Manual

That just sort of snuck up on me, one day I realized I had created a Best Practices Manual. So, I focused on configuring my web site to serve that purpose. Previously I had both Resilient Modeling Strategy and Resilient Assembly Strategy, But I combined them into Resilient Modeling Strategy. I also changed the web site name from ResilientModeling.com to LearnRMS.com to accommodate the users that are spelling challenged.

A Best Practices Manual needs to be available to the user wherever he is and the perfect vehicle for this is a web app. if you go to www.LearnRMS.com using Chrome on a mobile device and select “Add to Home Screen”, it will create a shortcut named “RMS Web Site”, add a RMS logo to the shortcut and lock the screen to landscape when you are using it.

Remember this is all CAD neutral, and will work with just about anything that has a feature tree. I have ebooks, seminars, classes, and consulting. So there are a lot of ways to approach this.

I’ve really shortened the process of becoming a power user.

========================================

I’ll write more on this topic later on.

12 Replies to “Resilient Modeling Matures – Interview with Dick Gebhard”

  1. I’m going to give this a whirl, but I’ve got to say, that learnrms.com website is a pretty big hurdle. There’s probably not a single visual arts layout, web interaction, or teaching pattern it adheres to. Good design is hard, and I have no doubt that real effort went into the site, but honestly I think a text file would be less distracting. You’ve got to know what you don’t know. I’m also disheartened by the kindle book being V5, but the website indicating the current version is V9. Aw well, maybe it’s still better than the wild west that is solidworks. Fingers crossed!

  2. I haven’t been able to my workplace to adopt Solid Edge, but I have been able to get them to adopt the basics of RMS. At our last company meeting the operations manager highlighted this improvement by comparing a machine re-design before and after we implemented the changes. Just modifying the travel in one axis took a week and a half on an old design, but only a few hours now.

    RMS has helped us address some of SolidWorks worst weaknesses. Maybe one day we’ll truly overcome SolidWorks greatest weakness, the lack of Synchronous, and adopt Solid Edge.

  3. I have to disagree with the following comment:
    “Since feature sequence and linking completely determine a model’s stability, an unstable model is 100% the fault of the designer.”
    Design is not a linear process it has loops due to the iterative nature of finding the best solution. If your CAD tool is not flexible enough to allow the designer to make unexpected changes then how is that the fault of the designer? I see it as fault with an inflexible software.

    1. Yes, well, of course Gebhard is being political. He doesn’t need to annoy anyone, as he’s trying to create a CAD agnostic method. There’s also a middle ground between it being the fault of the software and the user. The software actually does offer a method to work correctly, but the business model doesn’t allow for it. Users are generally taught to play fast and loose with the tools – aiming for ease of use and speed of first creation rather than stability of models through changes. Mr. Gebhard is trying to put people back on the right track.

  4. Hey Matt-

    When using SOLIDWORKS, I’ve become almost exclusively a Top Down Assembly modeler. Love it. Yes I allow multiple external references. When the design is completed, I have no Assembly Mates to screw around with, Yeah!!

    Then, I’m free to break all the external references, if I so desire, which is true about 95% of the time. Nice to have this choice.

    BYT I’m using Fusion 360 more and more often these days, what a value it is. It’s very capable, has all the tools required at a fantastic price.

    I hope all is well with you, old friend.

    Cheers,
    Devon Sowell

    1. Hey, Devon, it’s been a while.
      I think if you used Synchronous, you’d see a big benefit in it, because there are no external references unless you really want them, but you can still make edits of one part and make a temporary reference to another part. Plus, you don’t have to worry about the feature tree, or sketches, or… list goes on. Somebody’s giving CAD users little blue pills, I think.

  5. “If people are reluctant to see the value in something like Synchronous Technology”
    Oddly enough that also includes most SE users and that baffles me to no end. The most powerful tool in their tool box and they don’t have time to learn it. I am not kidding that is the largest by far reason for why the majority of SE users I have talked to do not use ST.

      1. Change is hard! Giving up “control” can be even harder. That is what I perceive from CAD users who encounter synchronous technology or concept of history-free modeling. They think they loose all control of the design. Which isn’t true at all.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.