BIM Asset Management Made Easy

MARK PRITCHARD: And welcome to this Autodesk University class, BIM Asset Management Made Easy. My name is Mark Pritchard. And I'll be your presenter for this session. So just a few formalities first before we get started, just having a look at our learning objectives.

So after this class, you will be able to define and create robust search workflows, build healthy libraries using BIM assets, and identify how UNIFI can be used throughout the process while collaborating productively across a global practice. In today's session. I'm going to be talking specifically about UNIFI. And for those of you who don't know, it was recently acquired by Autodesk earlier this year.

Little disclaimer, this is just our story about how we overcame the obstacles of BIM asset management. We think it was a successful story. So therefore, I'd like to share that with you today. So I'm not going to tell you that this is the way that you have to do it. It might not necessarily work for you and your firm. But hopefully, you will be inspired by the story.

It's a little bit about myself. I was born in Africa and grew up in Australia. I've lived in four different continents over my lifetime. In 2007, I moved to London where I joined HOK as a BIM coordinator, worked in a variety of different projects throughout my time there. In 2014, I relocated to our New York office, where I've now taken on the current role that I'm in at the moment.

And during this time, I worked on one of the biggest projects in my career, which is one 1.3 million square feet, little tiny little airport in the Northeast of New York City. Some of you may know it.

In 2021, transferred back to London to be closer to family. And just a little fun fact, I was a competitive swimmer at one point. And I've estimated I've swum about 62,000 miles over my career. So

I like to travel. I love playing and watching sports. As I mentioned a second ago, I'm an amateur baker I guess. I like to dabble a bit. Also, not afraid to have a bit of fun and dress up every now and then just to get a bit of a laugh.

A little bit about HOK-- we were formed in 1955 by George Hellmuth, Gyo Obata, and George Kassabaum. We have about 1,800 people across the globe in 27 offices or so across three different continents. We're a global design firm. We specialize in architecture and interiors, engineering landscape, urban design planning, visual communications, a few different market sectors. And we cover about 10 different disciplines as well-- so like I said, education, sports, government, and all that kind of stuff.

We have a small team of firm-wide directors and specialists who set the overall strategy for the design technology and establish the standards across the firm. We interface with our local design technology managers who then are responsible for implementing those standards and procedures. We have an annual summit every year that we get together. And this is just an excuse to have a good time, meet up face to face, and just exchange some general information about BIM and practices.

There's a few notable projects that HOK has worked on over the few years, just so happened to be some of the ones that I've personally worked on. But just to give you an idea, there's quite a broad range there from the UK, Saudi Arabia, Kuala Lumpur, Azerbaijan, a few other fun places.

OK. So let's get into it. So HOK has been a long time implementer of BIM. And this timeline is basically just a short history of HOK's BIM implementation. So back in 2000, 2002, like everybody else we were using Autodesk Architectural Desktop or ADT. And in about 2004, we started researching Revit. And then shortly after that, we implemented our first Revit project.

In 2008, we made the official change with our-- we dubbed it our HOK BIM day, which was the official date that Revit became our preferred platform. And then in 2010, we started introducing BIM certification to our projects. 2012, we started looking at a few different various content management tools like Kiwi Codes' Family Browser, and CTC BIM list, which later became their hive tool.

Although, these were actually successful, they did have some limitations, which I'm going to touch on a little bit later on. 2016, we then purchased UNIFI. We liked it as a library management tool. And the fact that they also offered a creation service was a plus. We messed around with it for a few months and then a pilot project before kicking the tires, before fully taking advantage of their content creation services.

And in 2019, we migrated all of our assets off of our servers and onto the UNIFI libraries. Then there was a little period where we worked at home for a little while. So we'll skip over that. In 2021, we made a big push to reorganize our libraries. And then in 2022, we had the added challenge of finding a new content creator.

Early this year, we've just been continuing to grow those libraries and just re-amalgamating stuff and cleaning up everything as we go. And then for those who don't know, UNIFI is actually going to be launching a new version later on at the end of 2023.

What challenges do we face? Like any other firm, we've got a centralized location for our digital library. This is synchronized through our Distributed File System, or DFS, to each office. But staff relied on our firm-wide team to maintain the network of folders. Versioning and updating of those assets was done manually, which meant that a lot of significant amount of time-consuming, tedious work.

Our biggest challenge was having assets scattered all across the firm. We had our digital library as well. But every office had their own individual library. We also had project-specific folders. We also had staff that were hoarding stuff on their desktops and their personal drives. And then of course, there's the internet as well.

Utilizing these various resources became problematic. When staff couldn't find what they needed in their local library, they'd resort to browsing past projects, reaching out to their local design technology manager, or even going to the internet. And this meant finding assets just took too long.

It also ran the risk of downloading content that was not necessarily 100% Revit, which resulted in model crashes and project delays as well. So collating everything in the local office and the project libraries took a bit of time as we set up collection folders where assets could be extracted to. Staff were also creating their own assets, as I mentioned. And that was also a challenge, because they'd save them to their hard drives or their local desktop. And once they leave the firm, it's extremely hard to get those assets back again.

And asset creation itself is an important part of the story. But it's a slightly separate spread. So I'll get into that a little bit later on when we talk about that. Once we began to gather up all of our assets, we discovered that we had a lack of good content, whether it was modeled by our staff or the vendors, it was either badly modeled large and family-sized, other embedded elements into it didn't meet standards, or there was just lots of duplicates of it. And I'll come back to this slide in a little bit later on. And we'll talk about this in a bit more detail.

I mentioned at the start that we previously used Kiwi Code's Family Browser and CTC BIM list. To be clear, Family Browser was extremely low-cost. And the cost of CTC BIM list was negligible since we already had we were already using other BIM suites of theirs.

But that still relied on assets being hosted on the digital library. And they didn't index other locations, which foundationally was why we like UNIFI. When we first looked at UNIFI, it was very cost prohibitive. In 2016, they actually revised their pricing model to make it a bit more flexible.

We also looked at Kinship and Avail as potential solutions. We didn't, however, look at the asset creation services. We just looked at them as a management tool. Our firm, my design technology team had now evaluated the options and decided to proceed with UNIFI. The pilot was our test of-- sorry-- the pilot was to test our assumptions and to make any adjustments necessary for the broader implementation.

So why did we choose UNIFI? UNIFI's a single-source, cloud-hosted platform allowing users to access it from anywhere in the world. It has a robust search engine. And automated Revit version updating and management made it a lot easier to deal with. At the time we chose UNIFI, they also offered an integrated service of content creation management.

Main feature of UNIFI is the libraries and the saved searches. This is fundamentally-important to good asset management. In this section, we'll look at how we utilize the service to its full potential. But first, a little background on UNIFI's libraries.

So UNIFI describes their libraries as buckets to store their assets. The libraries have their own admins who can monitor and asset-- monitor which assets get uploaded. And any users with access rights to the libraries will be able to browse and search that library to find the assets within them. When you search the UNIFI libraries, you can search either everything or you can search specific libraries as well.

And you have an unlimited, so you can create an unlimited amount of libraries. But simplicity is the key. Less is more. Keeping the library strategy simple will make it easier for users and also for your managers.

Early on, our initial library setup consisted of just two libraries, one for details, and one for general content. We uploaded some assets for the team to use in 2018. And then we added a metric library as well as our content creation library, as well as a content request library I should say after migrating our assets off of our servers onto UNIFI, we began to add more libraries to cater for some more different disciplines and a few larger projects.

A lot of staff were requesting project-specific libraries. But we avoided creating these for every single project. If we were to take a conservative average of 50 projects per year, multiply that by our 2,627 offices, we'd have around about 1,300 projects per year. And if you multiply that by the last eight years, we're just shy of 10,000 and a half, 10,500 projects. If we had a separate lobby for each one of those, this would be way too difficult to manage.

And besides, the whole point of having one solution with all the assets was to make it easy to find them. If we had individual libraries for each project, we might as well just left them on our servers. So instead, we created user libraries or user upload libraries where we encourage all of our staff to share the assets that they had procured, downloaded, or created themselves. This way the teams could share the wealth, especially if there were assets that were going to reuse repeatedly on future projects.

These lotteries were separate from the company libraries, which allowed the firm-wide team to observe what content was being used on projects. We could then check the modeling specifications on the families and make sure that they met company guidelines and were generally usable. Once they met these requirements, the assets were then pushed up into the libraries-- into the company libraries for further use.

Cut to four years later. And we had about 34 libraries. These were being named as we created them. So that meant they weren't very logical hierarchy. This meant that it was confusing for staff to often find the assets in the right place. They were searching the wrong locations. Or there's just libraries that we didn't really want them poking around in. So therefore, it was important that we reorganize these libraries to be more logical.

So we started off with our company libraries at the top, followed by our regional libraries like our Canadian and Asia-Pacific libraries. We also had some discipline libraries and a handful of project libraries. And these were mainly just large-scale projects that we were still extracting content from.

And then, of course, finally, we have our admin libraries where all of the admin design technology staff can poke around and set things up. Separating imperial and metric content into distinct libraries also helped reduce the amount of content that users were searching through. Having firm-wide shared libraries and regional libraries also helps separate the content for local jurisdiction reasons, especially for an international firm like HOK.

The fewer total results means the faster the designers can find exactly what they were looking for. This reorganization helped with a much cleaner, more logical hierarchy and helped direct staff to the correct locations for their assets.

Over the next couple of years, we continued to build more libraries that were basically-- sorry-- were focused on disciplines. I mentioned at the start of the call that HOK was a global design firm with fingers in every market. We had aviation, health care, lighting, MED planning, to name a few. And each one of these teams wanted their own specific-- sorry-- each one of these teams uses their own specific assets. They didn't want to have to sift through all of the architectural stuff to find their assets that they particularly used.

So we worked with the teams to create their own libraries and set up their own specific discipline folders, discipline libraries. And then we migrated everything from our company library into those discipline libraries.

With the admin settings in UNIFI, the ability to set and assign user permissions to the libraries was a big feature. This allowed us to grant or restrict access to users groups to specific libraries. We can grant edit rights to the design technology team who can then manage the libraries and their assets within it. It also meant there was no risk of users potentially causing any harm or removing stuff accidentally.

At the time of this recording, this is how our library growth looks as of today. Early on this year, we removed a few of our older product libraries that were now redundant and didn't have any assets in them. And this is now roughly an even split between imperial, metric, and some mixed-use library folders.

UNIFI is not just a management for Revit families. It can also host other file types, for example AutoCAD files, PDF, or image files that are used for applying to materials. It can also hosts text files for type catalogs, IES files for lightning families and even fill pattern files, but also, Revit schedules, drafting views, and, of course, Revit content folders-- I'm sorry-- Revit RTV files for content container files.

This has been extremely helpful for us when it comes to our detailed template library on our internal hub. UNIFI was a good vehicle to organize these details. And this has been uploading-- these details have been uploaded as individual drafting views as well as whole sheets of details and complete container files with multiple sheets. And this allows the users the choice of which one they would prefer to download.

When we disengaged our digital library, we had just over 8,000 assets. At the time of this recording, we now have about 29,000 pieces of content across all of our libraries. And that's roughly 80/20 split between imperial and metric.

Now, if all of those keen mathematicians out there, you'll notice that the figures on these columns do not quite add up to the total shown in the red there. The figure in red shows the total number of assets across all of our libraries. And this is by unique file name count.

The number of assets, however, are hosted in-- sorry-- a number of these assets, however, are hosted in multiple libraries for whatever reason. So the figure in the chart is a tallied count of each of these individual libraries. So basically, this exercise has told me that I still have a lot of clean up to do when it gets back to these libraries.

Now, that we have all these assets in one centralized location, how do we search through those assets to find them quickly? Before we get into that, let me just explain how the saved searches within UNIFI work. Saved searches are intelligent queries that can be used to group content. These are a powerful way to quickly and easily access particular sets of content for the design teams.

Saved searches can be set up by library or by company admins and are instantly available to all users. There is effortless management when new assets are added to the libraries. Because assets that match the saved search filters automatically get added to that search.

Saved searches will hide the complexity of advanced queries from users as a single click gets access to the content within seconds. And this feature is just as important as the libraries for your overall strategy.

At the same time as we're establishing our libraries, we also needed this way to quickly and easily search through those libraries to find the assets for the teams that they needed. UNIFI's search functionality increases efficiency for project teams by allowing users to create predefined search criteria. This prevents users from having to type in complex filters or remember exactly where to find the content and browsing through the libraries and the model categories.

Using UNIFI's meta tags and tagging the assets appropriately, we can begin to build these saved searches. However, instinct will tell you that a furniture asset should be tagged as furniture or as plumbing, et cetera, which meant that this quickly appeared to be too broad of a search term. So we had to drill down and get more specific in the tags.

Mimicking our company's family farm naming standards, we began to tag the assets with a bit more of a logical manner. So that they tied better into the family name, for example, lounge chair, bar stool, et cetera. Specificity being the key word.

Even though users have the ability to create their own saved searches, we did not want staff going crazy and making searches for everything. So through UNIFI's admin settings, again, we're able to control this and limit the creation of content searches to just the design technology managers only. Users can still create their own personal saved searches, but these are not visible to the rest of the firm.

So once the saved search group is created, we begin to build out and develop the specific searches, like I said earlier, the furniture ones. Then when the teams began using these searches to find their assets, we started finding the holes or more likely the roadblocks that were preventing teams from finding the assets faster. So I worked with individual discipline teams to build more appropriate saved searches that were more specific.

Of course, the majority of them were built for imperial assets as that was the majority of our library not wanting to forget about our colleagues in Canada and the UK and Asia-Pacific. We split the already created saved searches to show both imperial and metric. And this, of course, meant tagging metric-related assets appropriately. With this method of creating saved searches, it meant that it very quickly the number of assets-- or sorry, the number of searches being created was increasing rapidly. And without meticulous organization, it could get out of control very quickly and become another mammoth task to constantly manage.

Now that our staff are beginning to explore the libraries and the saved searches, we asked ourselves how do they engage with the assets? Designers are visual people after all. So we built container files and began exploring different methods of how the staff could access these container files to view the content within them. So we looked at Enscape, Twinmotion, and even Speckle and a few other softwares as well.

So let's go back to the slide that I mentioned earlier, our challenges. So I mentioned that we were having when we started collecting all of our assets, we found a lot of it was bad content. So it was badly modeled. It was either over-modeled or it was just line work. They either had no parameters or they had way too many parameters.

They were quite often very large and family-sized, as you can see at the bottom in the file size. They often had other embedded elements. There were still a lot of AutoCAD-- 3D AutoCAD, SketchUp, other softwares that was embedded within the Revit families. And they were often not very BIM friendly.

A lot of it didn't meet the standards as well. And there was also duplicates due to the fact that every HOK office was creating their own assets. We did an exercise where we searched a handful of our biggest projects across a firm. And we found that we had multiples of the same family in various forms across those projects.

So let's dig a little further and look at the cost implication or the cost impact of building those families. This table shows elements that we consider crucial when building cost-effective assets, user, or staff creation was high impact due to a few factors such as billable rates, staff skill levels, how quickly and efficiently they can build their families, and whether it meets the firm's standards and gets shared across the teams.

Compare that to outsource creation, and you can see that this is a much more cost-effective. Our firm-wide team has experience with outsourcing. We have always considered an outsource creation. But it has always been a project-specific-- in other words, assets that were specifically requested by projects and the cost was picked up by that project. And of course, it then never gets shared with other teams, we never really considered outsourcing on a global scale.

So this graph was based on actual numbers, which I'm not going to disclose today. But the firm looked at it and recognized the savings. Why spend money when you can have the right company built it once? And then we thought what if-- what if we didn't create assets anymore? What if we asked our staff to stop making BIM assets?

There had to be a few caveats though. Staff would need to plan ahead and make-- and have an idea of what they require. They can't leave it until the last minute to request a family to be built and expect it the next day. That would also need to familiarize themselves with the correct process and what would be required.

And then that would be covered at some initial training at the outset. The design technology team as well would also require some management and the process to make sure that things maintain momentum. As part of that management process, we developed a flowchart as a graphical representation for the process to follow when submitting a request. This is a simplified version of that chart.

Effectively, there are five main steps. The user submits a request. The asset manager triages that request to make sure all the correct information has been submitted. The vendor then creates the draft version of that. And then the asset manager, the design technology manager, and the user will then review that draft content to make sure it meets all the requirements that they initially requested.

If if doesn't, it gets sent back to the vendor to recreate and pick up any deficiencies, and then once the user is happy with that, it gets approved and then pushed up into the libraries on UNIFI.

But someone needed to manage that process. At the time, it was a revolving role amongst five of us on the firm-wide team where basically anyone who had time would review the requests. This is where I came on board. And although, I was still 50/50 on projects at the time, I took on the role and moved into it full time.

As recreation is a valuable service for our project teams, especially for those who are on a tight budget. By outsourcing the creation of custom families and consolidating and making those families available to all teams to use, duplication of effort is avoided. And the labor of creating Revit families will not have to be charged a project.

This meant that teams could continue to work on the design maintaining velocity and focus while ensuring BIM content is not recreated many times over finally reducing the workload on the overhead team on the firm-wide team.

So like our process chart, we developed another flow chart for sourcing these assets. And again, this is a simplified version of that. So the first step would be search the UNIFI libraries, make sure that family does not already exist. If not, we can check the manufacturer's websites to see if they have any downloadable families.

For many years, HOK's policy was to discourage downloaded assets for Revit. However, as more manufacturers are making their products available in BIM, the quality of the downloaded content is increasing. And some content can be reasonable for reuse. However, we still strongly encourage staff to follow the guidelines to evaluate those assets before loading them into their projects, which leads us to our next step.

These families need to be interrogated by our Revit model guidelines to make sure that they are usable and BIM-friendly. If the manufacturer doesn't have any content, then we can go back to submitting a content request to our vendor. And then once that request is approved, it gets uploaded into the library.

So by outsourcing the creation of most custom families and consolidating the slot, the HOK teams-- the duplication of effort is avoided and labor of creating these families is not charged to the project. So if we step back a little bit and we take a look at the amount of assets that were built for us by UNIFI over the last four years with their content request service, you can see the trend line shows an expected downward curve.

Lots of requests were submitted in the first year. And it gradually tapered off as content needs were filled out. The figure there of 893 is probably slightly off as a few of these requests did result in multiple families. But this gives you a good idea of how many we've had created.

By 2022, everything was going great. We'd survived two years of work-at-home libraries and saved searches we'd been built and modified to generate the most efficiency. And content requests were being submitted regularly.

But then we received a notification from UNIFI that they were going to cease their content creation services that we'd been taking full advantage for four years. I'm not entirely sure of the reasons why. UNIFI could probably best answer that. But with the announcement that they were building a more robust version, it's probably safe to say that they were trying to free up resources to build the platform.

Not wanting to forfeit this kind of service, we began to search for a new vendor. And thanks to a Google search, we began to build a list of vendors and then reached out to them individually to start having conversations about what services they could provide. After eliminating the not-so-good ones, these are some of the ones that we ended up contacting for quotes. We did our due diligence and dug a bit deeper into what they could provide and narrowed it down to these five whom we began having more in-depth conversations with.

Each vendor had its pros and cons. I'm not going to go into the specifics of what each firm offers, because that's beyond the scope of today's talk. But I'll touch briefly on what we looked for. We put them through their paces to test the quality and speed of the content creation much like we did with UNIFI a few years ago, I chose some families as mock requests. And these were families that UNIFI had struggled with for various reasons.

I wanted to really push the boundaries on what they could do. So I gave them very little information about some of the test families. Basically, I was trying to mimic what our users would submit as a content request. For example, I only gave one of them a PDF. The other one I gave them just an AutoCAD file and none of our BIM standards or shared parameters.

And I did this because I wanted to see how they dealt with the information. Did they ask any questions? Or did they just make any assumptions? We then cross-referenced that with the additional services against the results of the mock tests and use a simple star rating system on the various vendors.

From this, we were able to determine two companies that were not only cost-effective, but also had most to offer, and not just in terms of a content creation point of view. And these were the firms that we decided to go with on a hybrid style approach. Both high-tech, digital, and Kinship are willing to be flexible in their contract structure, which allowed us to use them both. So we could capitalize on the strengths of both firms.

High-tech digital would be our to day-to-day content creation service, which staff would request content from. We've also been talking to them since then about additional BIM consultancy work. Kinship is a firm that we request more specific Bespoke content from-- large suites of families, for example, wayfinding signage for an airport project. This would be the content that teams are happy to take their time with to get correct and exact. Kinship also has additional services a lot like UNIFI in terms of library management.

So let's take a look back at the assets we had built from high-tech digital over the last 12 months. There's been an initial upward trend, which we were expecting. And we'll probably see a-- starting to see a downward trend towards the next 12 months. Again, these figures are slightly higher in reality. Because some of the requests did produce multiple families.

So what's next? You might have heard this already. But UNIFI is planning to launch a new platform, the version four platform. Here are some highlights of what's coming. New customizable results, some regional deployments, new 3D previews, which allow you to see your content before loading into the model, a redesigned search filter, and integrated with other softwares like Civil 3D, Rhino, and AutoCAD.

HOK has been one of a handful of firms that have been testing this new version with UNIFY over the last 12 months. HOK has also been on the advisory board of UNIFI for some time now. And I have to say it's very refreshing to be on an advisory board where we actually feel listened to.

UNIFI's a great team-- Dwayne Miller, the CEO, Laura Higley, Scott Shiple, and the marketing team, Karen Pierce, Dan Reed, Tucker Cowie, and the rest of the development team. I'm sure there's others that I'm forgetting. So I apologize to those who I've missed. But they're a great team to work with. And if you want to know more about UNIFI's products, feel free to contact them directly.

So to summarize, we had boom assets scattered all over our firm with lots of duplicates. And we were manually collecting and upgrading which was very time-consuming. UNIFI's automated versioning and upgrading saved a lot of time and effort. Being a single-source, cloud-based tool we're able to consolidate all our connection points when we migrated our assets to the UNIFI library.

Developing a logical hierarchy allowed us to monitor those assets that were being uploaded and ensure they were going to the correct libraries. And using the powerful saved searches, we were able to target specific groups of assets for the teams making it more efficient and saving time.

The ability of UNIFI to store other different file types negates the needs for separate library locations for assets that were not Revit-based. When searching for a new vendor to create content for you, there's a lot of choice. So consider the turnaround time for creation, the quality of the final product, and, of course, the cost impact of outsourcing versus in-house modeling.

Setting expectations for your staff, your management team, and the vendor will ensure that all parties understand their respective obligations coupled with some training on the tool will help to confirm those expectations. And finally, UNIFI's a great team to work with. They're very responsive to any feedback. And they're always go above and beyond when it comes to their customers.

Thank you for listening to my Autodesk presentation today. If you have any questions, feel free to reach out on LinkedIn or via the Autodesk University website. And I'll be happy to help any way I can.