Connection Design Workflows Using Revit, Robot Structural Analysis, and IDEA StatiCa Connection

RALPH PULLINGER: Hello and welcome. This is presentation BES2691, Connection, Design, Workflows. This is what we're going to be looking at today. Some brief introductions about myself and the company I work for, a bit about the background, and the problems I and my company have seen over the years. I will introduce a workflow that will work in any situation with practical examples. And last of all, there will be time for questions.

My name is Ralph Pullinger. I've been a structural engineer for over 40 years. I was born in Birmingham, England, hence the strange accent. Following a successful stint in consultancy, I have spent the last 15 years in the software industry helping engineers, detailers, and management realize their potential. My favorite question to ask is why, and you'll find out why later.

I worked for a company called Idea Statica. We are based in Brno, in the Czech Republic. It would be interesting to see how many of you have heard of us. We are a global software company specializing in both steel and concrete solutions with over 100 employees and 40 plus resellers. We are a trusted technology partner with Autodesk, to name but one. Our software is used extensively around the world and our results have been validated and continue to be validated by several renowned international universities at the top of their fields.

A bit about the background of the problem we are looking at today. One size definitely does not fit all. Methodologies for connection design exist that cover most countries. Some processes are more complicated than others. Over on the West Coast of the US and South America, we see what we call the structural bundle or code of standard practice, option one or two.

As we move steadily East, we see what's termed delegated design entering into the Eastern US, or what is known locally as code of standard practice option through A, B, or C. This delegated design even hits the UK, back home.

Over in inland Europe and Australasia, we tend to see an all in one approach. And all of these approaches, I'll go through in greater detail in the coming slides. Some of the gray areas we don't want to know about or have just not been able to assess. The design options that we're going to look at relate primarily to standard contracts. But when we see an IPD contract, things change up a gear or two.

So the first methodology, the structural bundle or cost one or two, can be found on the US West Coast. Although, it might be known locally under different names or even a different name. There is obvious benefit to be had with all of the design being done by one organization. The drawback, however, is that there is little or no fabricator involvement, which can lead to uneconomical use of material and sections.

The second methodology, known as delegated design, is perhaps the one where there is most to gain. Here the fabricator assumes responsibility for the connection design using information gleaned from the global structural design, but provided by the engineer. That is the key differentiator. The potential for over-design grows if valuable design information is not shared adequately.

The third methodology, which we've called, all in one, is very popular on mainland Europe and Australasia. Having seen this first hand over a number of years, I can see the obvious benefits. Coordination being the obvious. However, this workflow could still benefit by moving from an analog to a digital basis, as far too much paper is still being used.

Over the years, we've seen issues arise that relate back to the load effects shared with the designer. These load effects were not based on bona fide combinations, but the maximums or envelope loads. And these loads, again, were factored up by a certain percentage. This approach dates back to when I was starting out as an engineer and is still prevalent today. What this means in reality, is over-design, as connections that should work invariably do not.

When it comes to checking, we know that engineers struggle with the information overload that happens in design. But imagine a scenario where calculations are issued together with a 3D model of the connection that can be interrogated and viewed from any angle. Communicating all this to the fabricator efficiently is the next hurdle to overcome. In order to do this, they should be able to receive the design intent to base their detailed modeling on. It should also be easily incorporated into their design. If we are able to create a flexible and reliable workflow, one where information is not duplicated, then our industry can look forward to a bright future. One where risk is lessened and time and cost overruns are a thing of the past.

Some of the reasons why information is not shared is often down to two letters. I and P. Either intellectual property or professional indemnity. The former is often linked with the exchange of whole models, which are not really required by the connection designer. If the relevant load effects from the actual load combinations are published, then there is no need to worry. Indemnity, or insurance, should not be an issue either, if these loads are taken from the same model the global design came from.

If the engineer is not prepared to guarantee these results, then what is the point of the design. Sharing markups, either as PDFs or even as drawings, offers no real benefit. It can be seen as a massive step backwards. Excel schedules, on the other hand, are marginally better if they contain the correct information.

We all know that ours is a complex world with complex products, projects even, involving lots of consultants. Early engineering and fabricator involvement has a proven track record. Just look at ITD. If we were looking at connection checking, how do engineers currently do this? Do they wade through pages and pages of calculations looking for key results? Or do they recreate the connection design? Either way, there should be an easier and more efficient way to do this. If all of the design information could be shared, then checking should be quite straightforward.

The final hurdle to overcome is getting the information to the fabricator for detailing. It's the last thing to be done, but it's always the first thing that is required. Unfortunately, this is the way it's always been done.

We've talked a lot to engineers around the world. We've talked to fabricators, and we've talked to software vendors like Autodesk. We realized very early on that a good workflow allows engineers to work the way they want to with the tools they want to use and the ones they are used to working with. This solution being put forward today could equally apply to other solutions in the same field. We can connect multiple parties to streamline the design process at the end of the day.

On this slide, are the key processes involved in arriving at a successful connection design. We'll go through them one by one. In step one, we are seeing an early structural model that has been constructed in Autodesk Revit. This model contains all the structural information in the form of structural beams, columns, beam systems, slabs, and braces. No connections as yet.

It also takes advantage of the analytical model. That has evolved and developed over a number of years. Hopefully everyone knows that this information is available and can be used downstream by the likes of Autodesk, Robot, structural analysis, professional. However, we all equally know that this is not the case. Most likely the analytical model is being created by a different person, possibly in tandem. The process of transferring the model from Revit to Robot is quite straightforward.

The result of the direct air exchange facility from Revit to Robot is an exact duplication of the Revit model. This model could have been constructed from scratch, but that would have taken a considerable amount of time and duplication of effort. In this model are the same members with the same connectivity as per the Revit model. Loads and load combinations have been added. As you can imagine, modern codes create a lot of combinations. And once we have a good model, we can analyze it and generate an even greater number of results.

Sometimes it's quite hard to visualize what these results mean. But thankfully, Robot allows us to investigate and sense check our inputs. We can check the results by visualizing moments. We can check deflections by choosing the most appropriate load case and enabling the deflections to make it easier to read, we can highlight certain areas of the structure and get a better representation of the results, making sure it is as we expect. The big question is, what comes next.

The next step would be to update the structural model with the results of the structural analysis and design. Using the direct approach, we can integrate the results as well as any geometrical changes, i.e. changes in member section size. This is accomplished very easily by the reverse process, i.e. we're importing the data from Robot rather than exporting it to Robot. And Revit being Revit, any changes to the model will instantly update. However, this is more likely to be accomplished by a series of mark-ups indicating the changes. And I've yet to see this workflow, this digital workflow, in full use. If we can update the Revit model, then the results can also be stored and utilized further downstream. The results are stored in the results package. And we can explore those using the tools within Revit and again, even use these to document the design. We have access to the full scope of the results that we have developed within the Robot model.

So we've now seen what can be undertaken by one organization or even one person. We now have a structural design that is complete with results that have been validated that can be used again. The next step is where it starts to get interesting from the perspective of connection design.

What we can now do is publish the geometry, beams, columns, braces, et cetera, and their load effects to a structured database via solution called Idea Statica Checkbot. With Checkbot, we are able to recreate all of the key information required to design the connections without the massive overhead of the whole structural model.

This means that the connection designer receives the right information at the right time. Furthermore, with today's data sharing solutions, this is a relatively painless exercise. In Checkbot we are able to visualize the load effects and the load connections into design sets. This greatly simplifies the connection design process as we can now focus on key connections rather than all connections.

The load effects that we pull in from the Robot model are many. In fact, they are complete. We have algorithms within Checkbot that will reduce these to the most onerous. And we call these the critical load effects. After all, if we've got 163 load cases, we don't really want to be wading through all of those when we can just design on a subset of about eight, say, for instance.

Visualizing the load effects is as simple as selecting the connections that we are interested in, or sets of connections, and drawing those load effects out on the screen. If we were to open a column base with a brace connection, we can see the load effects that have come from the global model.

The load effects, as I mentioned, are many. The load effects have been reduced further to a subset of about eight. And we can see each load combination being applied to the model in turn. So when we design this connection, we don't design on maximums. We design through each and every load case. As a side note to this, once we have a connection, Idea Statica is also able to conduct a stiffness analysis. So the usual engineering, fixed or pinned assumptions can be validated and improved upon for a high level analysis check.

At the same time, fabrication details will start to emerge. Again, we are reverting back to Revit to show how these could be applied. These connection details may or may not be adequate, depending on the skill and expertise of the detailer. If we were to add a series of simple shear tabs to one of the top beam connections, we can visualize those quite easily in Revit. But we really want to be able to check these out in a more substantial manner.

Revit has plenty of tools that will comply different types of connections from simple shear plates to complex moment connections. It can apply these to different cross-sections with different arrangements. But what is relevant is that this model is using the same information that is also in Robot. As an additional step, not shown, it is also possible to export the steel work in its entirety to advance steel.

We've seen how to get the information published from Robot. What we're going to do now is publish a different set of information from Revit. In essence, it's exactly the same process. We use Checkbot to create a database based on, let's call it the fabrication model. And we are lucky in this case, that the load effects are also present in this model, so we don't have to ask for them.

If we were working with our results, say by using another analysis solution or if we were using Advance Steel instead of Revit, then we also have the option to merge geometry and load effects for the same connection across two databases. We can also request one of those intelligent Excel schedules, as we are able to import the data directly from Excel. But the important thing to recognize is that the schedule needs to have the right information.

This is one of the ways that we can break down the silos associated with connection design. If we run this initial connection through our comprehensive checking solution, Idea Statica connection, we will see that it fails, and it fails considerably. These checks that we are doing conform to the AISC code, but there are other codes that we can use in the design of our structural frames and connections associated.

When I said considerably, I meant considerably. If we see red on the overall check, red is bad. Luckily, we are able to investigate these shortcomings and assign a design from over 700,000 valid examples we have on file. When we run this through the check, the results are a lot better. If we see green, we're seeing sections that are being utilized well. If we see gray, they are not being used to their full extent. Luckily, we see no orange, which is bordering on bad, or red, which is bordering on-- which is definitely bad.

Now, unfortunately, we now must communicate these changes to the detailer. And previously this was done by sketches and change requests. Now we have change requests and access to a high fidelity IFC model that can be read into an IFC viewer or indeed read back into Revit. From a checking perspective, we can create an online model that is the same as the one created on the connection designer's desktop. So not only can we issue them the actual calculations but also a link to this 3D viewer.

This functionality can be further enhanced if the checking engineer also has a valid license of Idea Statica. It's as simple as sending an email. If you can type. So what have we just seen?

In step one, we created the initial structural geometrical model. In step two, we created the initial structural analytical model, which we could do either directly or independently. We chose Robot for this task. In step three, we updated the Revit structural geometrical model. And in this case, we also elected to pass the results. In step four, we created a structured database of connections with the geometry and the results from the Revit model.

This is the database that could be shared by connection, by two connection designers. In step five, we carried on and we created those fabrication level details. But imagine a model with many more connection details. How would we go about checking those connections one after the other?

And in step six, we created a database that could be used to design and code check those same connections that we created in Revit. The connections in steps four and six utilized the software. From Idea Statica.

In summary, then, I've shown you how to-- how connections could and should be designed and code checked. By reusing the right information at the right time, by the right people, this process will enhance your workflows. This workflow, or something very similar, can streamline your overall processes, reduce your risks, and create more cost effective designs that, by direct comparison, will also have less embodied carbon. A cheaper design has less carbon.

In reality, we can work with virtually any analysis, solution and any BIM solution. Just look at who we're able to partner with. For ease. I've highlighted the Autodesk Solutions with the red box. The process for many others is almost exactly the same or very similar. And I just go back to what I said in one of the original slides is that we empower engineers to work with solutions that they want to work with.

Thank you for being here.