AU Class
AU Class
class - AU

A Digital Revolution in Resilient Housing, Build Change, and Autodesk

,
共享此课程
在视频、演示文稿幻灯片和讲义中搜索关键字:

说明

From designing retrofits in AutoCAD to automating Revit workflows using Dynamo, and now exploring the many uses of Forge, Build Change has been using Autodesk products for 17 years to revolutionize the design of resilient infrastructure, particularly in the housing sector, and successfully reach national scale in many countries. By leveraging the latest Autodesk tools, Build Change is making home retrofits safe and possible in the COVID-19 era. In this session, we will explore Build Change’s 17-year history of using Autodesk products to impact 600,000 lives across 24 countries, providing an integrated solution for improving the resiliency of homes and communities and preventing loss of life in natural disasters. This session will feature a case study of the work Build Change is doing in Colombia, one of our largest country programs, and the Autodesk technology being used to scale housing retrofits there.

主要学习内容

  • Learn how Autodesk products can impact infrastructure resilience, particularly in housing in emerging markets
  • Learn how to implement digital solutions for infrastructure resilience
  • Learn how to develop flexible, adaptable BIM tools for many different use cases
  • Learn how to solve bottlenecks in the construction value chain using digital tools

讲师

Video Player is loading.
Current Time 0:00
Duration 0:00
Loaded: 0%
Stream Type LIVE
Remaining Time 0:00
 
1x
  • Chapters
  • descriptions off, selected
  • subtitles off, selected
      Transcript

      ALLIE YOUNG: Hi, everyone and welcome. We're happy to be back at Autodesk University again this year with our session called a Digital Revolution in Resilient Housing Build Change and Autodesk. But before we get into the presentation, we're going to take a minute to introduce ourselves.

      So my name is Allie Young. I'm a Civil engineer from Canada and I'm a BIM technology specialist at Build Change I've been at Build Change for over a year and have supported Build Change programs in Nepal and Colombia.

      I'm here today with my colleague Andres, who will introduce himself now.

      ANDRES ROBLES: Hey, thank you for joining us. My name is Andres. I'm also a Civil Engineer and have been with Build Change for over a year. I've been supporting the Colombia program in BIM, and engineering topics. It's a pleasure to be here.

      ALLIE YOUNG: Great. So our team has presented at Autodesk University for a few years. So if you've seen one of our presentations before, you might be familiar with us but just for some background Build Change is a non-profit organization that was founded in 2004 by Dr. Elizabeth Hausler, with the mission to save lives in earthquakes and other natural disasters. Our aim is to greatly reduce death, injuries, and economic losses caused by housing and school collapses, due to earthquakes and other natural disasters in the global south.

      So we are currently operating in about six countries, including the Philippines, Indonesia, and Colombia, and have previously worked in many other countries, including China and Haiti. So here's a map of some of our current and past locations where we've worked. Now this is our favorite chart to use in almost every presentation. But I think it's a great way to show the importance of the work that Build Change does, because substandard housing is a growing problem in the world that, according to the World Bank, will affect three billion people by 2030, which is a massive number of people.

      So substandard housing can be broadly defined as housing that is inadequate and unfit for human habitation, mainly because it does not meet building standards and codes. So here are some photos from the communities we work in, including the photo on the right, which is a pretty stunning photo of informal housing in the south of Bogota, Colombia. It really shows you the scale of it.

      But not only is the issue of substandard housing a problem in itself, but it will only be compounded in the future by the climate crisis, which is increasingly bringing more severe weather to our cities in the form of storms and floods, and requires our infrastructure to be more resilient. So in many of the countries we work in where informal housing is prevalent, you can see how this could be a huge problem.

      Since 2004, we pursued the mission of safeguarding lives in natural disasters through many ways, including training builders in earthquake resilient design, and creating educational resources on safe building. But in terms of actual construction, Build Change has used two main approaches. First, we have a new construction approach [AUDIO OUT]

      at a collection through design phases, which we are going to focus on in this presentation, and which have gone through several automations over the years

      From data collection to design, our biggest goal is to generate retrofit designs tailored to each house but do it as quickly as possible. This includes first, effectively and efficiently, gathering enough data about the house to generate a design, storing and sharing that data with all stakeholders, then developing the final retrofit design tailored to that house. So this process has changed a lot over the years and in the various countries that we worked in.

      So Build Change has worked with Autodesk software since the beginning. To begin with, the team is primarily designing new construction projects using AutoCAD, such as this house in Indonesia. Many of these designs were created for post-disaster scenarios. And at this time, using AutoCAD was obviously standard in the industry, as BIM was just beginning to emerge.

      So here we actually have some examples of old drawings from Indonesia in 2006. Back then Build Change's methodology was to design specific type designs, such as this one, and replicate them over and over again in post-disaster areas for as many families as possible. Also we have some examples of drawings from China in 2009, when Build Change supported the construction of new safe houses there. For many years it was standard for us to create designs and details in AutoCAD, slightly modifying the designs each time to suit the specific needs of a homeowner or the specific size and shape of a lot.

      So our shift away from a completely AutoCAD based workflow started to occur in about 2015 when Revit was first introduced to the Haiti program. So in Haiti, an Autodesk's student expert intern helped to create our very first Revit template, which at the time was a really big change for the team and faced a lot of issues with adoption. But this marked the beginning of Build Change's work with BIM.

      So our Haiti template was followed by a new Revit template in the Columbia program in 2017, made by another Autodesk intern, that was designed to streamline our workflows and make them more efficient for retrofit projects specifically. So here are some examples of our first Revit models from the template in Haiti, which was used to model type designs for new housing construction projects. So here we have a 3D view of the foundation and its reinforcement. And also a 3D section of one specific type design.

      So while these Revit models were amazing, this was still a basic and un-automated implementation of Revit and there was a lot more to come in the future for us. The next really big leap in terms of adopting new technologies came in our Nepal program, which at the time was using a few type designs drawn in AutoCAD for their new construction projects and replicating those designs over and over again. However, with more support from Autodesk consultants, Build Change was introduced to the automation power of Dynamo for the very first time. With Dynamo, Build Change realize the potential for automation in designing retrofits, which were previously done on a case by case basis, because every house is obviously unique.

      So here they created the first workflows that leverage Dynamo to automate retrofit type designs, which were retrofit designs replicated over and over again on similar housing types. This was Build Change's first glimpse at how they could actually reach scale and doing retrofits by using Dynamo. So this is the point where methodology meets technology, because Build Change adapted their methodologies for retrofitting so that the retrofitting process could be easily performed using Dynamo.

      So for some context, I have this video of a virtual tour of a retrofitted house in Nepal. But in rural Nepal, the dominant housing typology is very prevalent, which means that many of the houses are similar and at times almost identical. The houses are made out of mud and stone, and are often rectangular in shape with one and a half to two and a half storeys. So because these houses were so similar, the retrofit designs could be similar too, much like the retrofit that was shown in this video.

      So Build Change, in addition to type designs for new construction, created type designs for retrofits. So for houses of a specific category that met certain criteria, such as rectangular 2.5 story houses, certain rule checks would be performed on the houses in Dynamo that led to specific retrofitting outcomes in the Revit model. So for example a Dynamo script would evaluate the length of a house and add special columns, called strong backs, to the perimeter of the house at a specific spacing. Using this method every step of the retrofit design process could be automated using Dynamo. So here is a video showing the first type design in Nepal and the structural elements that were being added to the houses automatically in Dynamo, including a ring beam and strongbacks anchored to each wall. So that's one type design.

      But next I'm going to show you a demo of another type design and one of the Dynamo workflows that I help to work on in 2019 for the Nepal program. And this is for a type design called the splint and bandage method, which is a very unique method of retrofitting. So in a splint and bandage method, reinforced plaster is added in bands to the top and bottom of openings, and in splints to the side of openings, to reinforce the openings. There's also reinforced plaster added to the corners of the building. And as you can see in the video here, buttress walls were actually added to the inside of the building as well.

      So this was one of the first fully automated Dynamo workflows that the team created, to make things almost automatic in Revit. But in order to retrofit these houses using Dynamo, we needed to also use Dynamo to import Revit models of houses from data collected in the field. So the team created a pretty basic method for generating a Revit model, based on filling simple forms with the houses dimensions.

      So a person in the field would measure the distance from every corner to the edge of each opening and input that information into a basic form, in addition to data on opening height and wall height. This form would output a CSV file that would be read by a series of Dynamo scripts to translate the data into Revit elements. So this was the first most basic form of our import to Revit process that has changed a lot over the years.

      But now I'm going to hand the presentation over to Andres, who will talk specifically about the evolution of our workflows in the context of the Columbia program, where there has been a lot of innovation recently.

      ANDRES ROBLES: Thank you Allie. Okay, so in this part of the presentation, we will take a look at case study for our current work in Colombia and see the technologies that we are using now and how has your workflow evolved over the years.

      So first, it's important to give some context on Colombia, in general, before diving into our work flows. Colombia is a country with very high rates of urbanization, with almost 80% of overall population living in big cities. This is one of the highest urbanization rates in the world nowadays. Unfortunately, a lot of migration to these cities has occurred on the periphery of the cities in the informal settlements where the families do not have permits to build and do not follow technical guidelines when they build. This presents a risk, because not only are the houses unsafe, but they are built in regions with seismic activity.

      On the map here, you can see that most big cities in Colombia, like Bogota, and Medellin, or in the Andean region where we have a seismic hazard. So the vulnerability of these houses, combined with the seismic hazard, creates a scenario that's high risk. Build Change has been working with government agencies here in Colombia. We're trying to address this issue, including the Ministry of Housing and Caja de la Vivienda, which means people's housing fund.

      Both agencies are helping subsidize improvements to people's homes in Colombia ranging from small habitability improvements, to full structural retrofits of houses. A good example of these efforts is enterprise's Safety Initiative, which is helping families to safely build their second stories on their homes. Build Change provides technical assistants to both these organizations to provide safer homes to thousands of families.

      So for the retrofit projects, there are three main steps we follow to generate an insider communication to start with the construction. First, the data collection phase, which in this context, means capturing the necessary information to create an existing plan of the house and accurately modeling the house in AutoCAD or Revit. Here, the second step is engineering, which consists in creating a retrofit proposal design based on Build Change's design manual performing various calculations on the house. And finally, the retrofit design is drawn into AutoCAD already and all the construction documents are delivered for review.

      During [INAUDIBLE] in Columbia the retrofit projects were done very manually, starting with data collection. The house plans were drawn by hand and then transferring to AutoCAD, and eventually into Revit, in 2017 when our first Revit template was created. This meant that the house needed to be drawn twice, first by hand and then by computer, to generate the existing plans.

      The engineer would then use a series of Excel sheets do calculations and come up with a retrofit design. This includes all the represented calculations, which compare the area of the house to the cross sectional area of the walls. This calculation is very important to change the engineering methodology, and previously was done completely by hand, which is a process that requires many iterations.

      The final engineered design can be modeled by hand in AutoCAD or Revit, where the final construction package will be generated. As an additional sub-step within the construction package, and in addition to Revit, the Revit generated the blueprints. The construction packages also require forms that deplete the existing state of the house and its elements, identifies deficiencies and clarifies how these deficiencies are there.

      Our workflow today automates many of these steps with the help of Autodesk team for tech our team was introduced to the magicplan in 2018. This is a third party app that allows the user to draw a detailed floorplan using a phone or a tablet. The purpose of this app is really to draw the floor plan only once and import that into Revit without needing to model it again.

      However, magicplan works particularly well for gathering spatial information and information related to the real elements of the house, which is why we also included Fulcrum into our workflow. Fulcrum is a third party app as well. And this app allows the creation of very detailed and adjustable forms to gather information. In this case, the information that we can gather is related to the house, its location, vulnerabilities, and also its homeowner. The information is especially important for filling the forms within the construction package, meaning it's a source of information that we can also gather and use automatically. We saw this as an opportunity to expand scalability within our workflow.

      But before I dig into this part, we must continue with the chronological order of our work, which continues with the magicplan projects being up on the server. From here, we begin the import process, which has also changed a lot over time. Originally a list of Dynamo scripts would rate an information file generated by the modeling application and construct the house piece by piece, starting with the project information, then adding the walls, windows, doors, foundation's, floors and so on. However, the user needed to get this file to a web page. [INAUDIBLE] we can say it was not very intuitive to use.

      Now we use much less processing power. And we use a much more straightforward and intuitive solution. The process is now involves only two scripts and can be done in under a minute. All the optimization tools we have acquired over time have been crucial to make these advances in represent nowadays the scalability of our developments. So it's like dynamic packages for user interfaces, for server connections, and also Python coding within the app.

      What you are seeing in this video is a summary of these scripts and the result of combining our tools in the most optimized way possible. The first script promises a user interface containing the names of all the projects available in the workspace, letting the user choose one. Subsequently, the second script reads this information and creates all the Revit elements found in the information file. You can see here on a side by side, that the model created on magicplan will be exactly replicated within Revit.

      Okay, so as explained previously in the context of our Nepal program, [INAUDIBLE] the same approach uses Dynamo scripts imports the house through a series of rule checks and then selects retrofitting outcomes based on these checks. In the context of the Columbia program, one of these checks is based on the percentage of a wall that is solid, versus the percentage of a wall that is composed of openings.

      The Dynamo script will evaluate this check and take action if necessary. In this case, the action would be partially filling up an opening, in this case a window. Another script will automatically add columns at wall intersections and columns to confine the doors in the house. These are just a couple of examples of retrofitting scripts that I run on the house in a certain order, in order to retrofit it according to the rule checks.

      For more scripts, we'll add a ring beam to the top of the walls adjusting the wall heights so that building can be airy. Then the free wall edges script extends the ring beam across any free wall edges to generate closed loops. These actions take place within the model and have an effect on the retrofitted proposal plans. If we run the entirety of this retrofitting workflow, we can end up with a compliant house under the guidelines of Build Change design manual, excluding the extra toll calculations, which will be executed afterwards and will depend on the configuration of the house.

      Okay, so finally, at the end of the workflow, we reached a list of the scripts that make useful calculations and also filling the forms that were once filled to Excel spreadsheets as a manual procedure. What these scripts all have in common is that rather than modify, they use the existing and the retrofit model to fill in the necessary tables for the construction package.

      As I mentioned before, these forms also take a significant input from the Fulcrum app. So we are using that to our advantage, automating a large part of the workflow in the effort of making the process of creating a construction package as quick and easy as possible for any given user. The first script out of these construction scripts, you'll see the images of the existing evaluation of the house. This document contains information about the initial state of the house and some key elements inside them. The images are retrieved directly from the Fulcrum workspace.

      Subsequently, the next four scripts focused on making the existing and retrofitted structural calculations and showing the user its results. That is for compliance purposes. This way the user may iterate the configuration of the house in the Revit model, in order to achieve compliance. What we can see here is a well represented calculation for existing retrofitting in our particular project. Finally, the last script builds in the form of deficiencies or as we call it LVD with information gathered from calculations, the finished model and also the field information that we can collect from Fulcrum.

      Okay, next we're going to talk about the future and look at some of the developments that we've been working on. The overall goal of the team is to reduce the time needed to retrofit a house, and more importantly, to reduce the technical expertise needed to retrofit houses. We want to remove as many barriers as possible that prevent the user in our work, so that people with little to no experience can successfully retrofit houses.

      To reach this goal, we want to make our workflow into a server, while also enhancing the user experience with an easier and more understanding user interface. That makes the user interact less with Revit. Introducing Forge, for the first time in our organization, we will be able to do just that.

      We are currently in the process of using Forge to move our input process into a cloud. So instead of importing the house in a desktop computer using Revit, the user will be able to send data directly from magicplan to a Revit model automatically uploaded import to this cloud service Named Autodesk Construction Cloud. We can achieve this by developing Revit plugins that do the same tasks as designing scripts.

      This mode to locate a Revit application server that executes the blueprints to Forge's design automation API already. So what this ultimately means is that we have to migrate visual programming logic maps into our coding environment, C# in this case, in addition to developing an application server. These tasks, as you can see, are headed to a very optimized coding environment, in which we are experimenting and discovering a whole new set of tools that we want to further use to improve the workflow.

      We hope to eventually move the entirety of our workflow into the cloud using Forge, which will allow the Revit model to be modified and retrofitted with less user interaction, increasing processing times, and increasing the scalability. And that's it for our presentation. Thank you so much for joining us. And we look forward to answering any questions you may have.