24 Jun, 2019

On Practicing Automation In AEC With Guido Maciocci

I made a few interviews over the past few months with folks from Dassault to offer readers an insight into how and why certain CATIA features came about and what they mean for end users. Changing lenses; today I interview one of CATIA’s experts from Australia, Guido Maciocci of AR-MA to talk about CATIA automation in practice. Let’s get to it!

1- What do you do at AR-MA?

AR-MA (Architectural Research – Material Applications) is a multidisciplinary architectural practice and virtual construction consultancy based in Sydney, Australia. The hyphenated name suggests the two aspects of the company. On one hand we operate as architects delivering innovative spatial experiences, on the other we collaborate with clients across the AEC supply chain to assist them in delivering highly complex architectural projects. The work is underpinned by technology, advanced computational methodologies, and a broad knowledge of industrial processes. In everything we do our main objective remains the same — changing the way we create the built environment by integrating downstream manufacturing constraints and the realities of construction into upstream creative design processes.

I have been part of AR-MA since 2012 and as an Associate my role is focused on high level problem solving, system design and detailing, workflow design, automation, and project management. Since we introduced 3DExperience CATIA into our toolkit around 5 years ago, I have also been responsible for developing tools and workflows, establishing best practices, and providing 3dx training to the rest of our team.

2- You have been using automation to support design to fabrication projects, what are the best and worst types of problems to automate and why?

We face a diverse set of challenges in our work depending on the type of project, the stage of the project in which we are engaged, and the requirements of the client. These criteria very much influence the kind of solutions we deliver but I like to distill the types of challenges where automation provides value into 3 categories: Time, Risk, Complexity.

In the first case we automate repetitive and/or time consuming tasks to reduce the time taken to execute laborious (but easily codifiable) tasks. We all work by the mantra “if you need to repeat the same task 5 times in a day you should [probably] automate it”. Good examples of this are creating nested folder structures for output files, renaming multiple files, automating title-block information, batch scripts for exports. We also develop scripts, tools and utilities to greatly speed up 3D modelling efforts such as instantiating UDF’s, Powercopies and Engineering Templates.

Secondly we automate to reduce risk. By creating automated systems and workflows where human intervention is limited or eliminated entirely, we greatly reduce the risk of human error. As a result our QA processes involve validating code, algorithms, and workflows, rather than individual outputs; the code either works or it doesn’t, and if it does we can rest assured that the quality of output is guaranteed. This is of particular importance when generating fabrication data for thousands of components, often times with a high degree of uniqueness.

Thirdly we automate to be able to solve complex problems. Design to production projects, particularly those characterized by geometrical complexity, often present challenges which would require an unacceptable amount of time to solve manually and some problems are just impossible to solve without some form of computation. Geometric rationalization of complex surfaces, planar penalization, cost reduction through material optimization, and material waste analysis are all great examples of these sorts of problems.

Our goal is ultimately to provide the most value to our client and automation enables us to deliver the best possible design solutions in a time and cost efficient way.

3- Which segments of your design to fabrication pipeline are automated? And what platforms do you interface with from 3DX?

Nearly every stage of our pipeline has been automated to some degree. Our 3D modelling, shop-drawing, and fabrication output workflows are heavily automated. Being a small company, we rely on our capacity to develop powerful custom tools that enable small teams to tackle large projects and perform the work of much larger teams in the same time or less.

We generally consider ourselves to be software agnostic, using the most appropriate tools for the task at hand. As a result we develop interoperability workflows that enable us to fully exploit our software stack. Other software we interface with 3DX includes Rhinoceros, Revit, Autocad, Navisworks, and Excel.

4-What languages do you use to automate 3DX and its integration with other tools? Can you comment on the viability of long term-big gain vs short term-quick wins aspects of the languages you use?

We make use of most of the available languages to develop automation in 3DX, including EKL, C#, VB.NET, VBScript. Each has specific advantages and applications.

EKL is an absolute must if you want to use CATIA with any degree of confidence and is fundamental to all CATIA workflows. It is a simple scripting language, easy to learn, and fast to deploy, yet it allows you to design incredibly powerful automation routines.

C# and VB on the other hand come into their own when designing any tools that require graphical user interfaces, or for interoperability between CATIA and other software. I have used C# to create tools that replace or augment OOTB functionalities, or provide tools that are otherwise unavailable (e.g. sequential object renaming, data extraction, real-time data visualization). They can also both be used to write Macros in CATIA which, unlike EKL, can be deployed as custom tools and commands.

Lastly VBScript is used to develop routines that can be called by end-users or programmatically by EKL Actions and Reactions when you need deeper interaction with CATIA; especially when there is a need to access functionality not exposed to EKL (object selection, database querying, interacting with the GUI, etc.).

5- What is your view on DFMA for AEC in general? Is the AEC market segment you’re serving receptive to DFMA? Is it still integrating VDC, BIM, or is it still 2D drawing affair? And how would automation assist in that?

While the construction industry remains overall a 19th century affair, I think that in the coming decade we will see a strong trend towards the digitalization of the AEC industry as a whole. There is really no excuse left for one of the biggest industries in the world to remain the most antiquated! DfMA is only one of the many puzzle pieces that need to fit together to create any meaningful transformation. I think true change will come from new forms of regulation, legislation, contractual models, and procurement methods that promote digital collaboration across AEC projects. I also believe that we need vertically integrated platform solutions that enable these collaborative processes; as opposed to the current situation characterised by slioed and fragmented data across different software and formats, desperately stitched together in federated Franken-models, riddled with mostly inconsistent “building information”.

Some years ago we would sometimes find it difficult to communicate the value of our work, too often reduced and misunderstood as “3d drafting”. Today we collaborate with some of the biggest names in the industry, and our work is a point of reference for best practices in digital workflows in AEC. AR-MA have been integral in the realisation of some really amazing architectural projects and every year we manage to reduce 2D documentation a little bit more, pushing for design, engineering, and manufacturing to be based on 3D models. We do this by trying to educate our clients as much as possible and delivering exceptional value to build confidence in these methodologies. It’s a slow process but it’s only a matter of time before the kind of work we do is the norm, and high fidelity, fully detailed, collaborative models of entire buildings will become required deliverables. I look forward to that day!

Tags: , ,

About : Maher Elkhaldi

Maher Elkhaldi is a senior applications engineer at Tesla Motors. He founded the 3DXAutomation blog to help make knowledge of programming CATIA easier to find, and contribute to the open-source community.

Leave a Reply

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