In this class, students will learn the basics of machine learning and how they can apply it to building design and construction. First, we'll talk about the history of machine learning and how it has been used in literature and the building industry. Then, we'll talk about some easy-to-use machine learning algorithms and try to implement them in Dynamo Studio software. This class does not require any prerequisite knowledge or skills.

Many industries have embraced machine learning, but its uses are still widely misunderstood. This class will demystify machine learning by providing a case study on how KLH Engineers used Python to create a machine-learning model to bridge the gap between AutoCAD software and Revit software in the architecture, engineering, and construction industry. This study will illustrate which concrete problems machine learning can solve, such as converting thousands of possible AutoCAD layer names into useable Revit elements by utilizing conversions already made by hand. Attendees will discover how to use their own historical data to enhance their company’s process workflow.

This roundtable discussion will get you up to speed on what machine learning is, what research is currently in development for Fusion 360 software, and how it impacts you. As Autodesk becomes a data company, machine learning can add value to your product development process in a variety of ways. We’ll start the discussion by giving practical examples of the types of problems that machine learning can help solve for product design and manufacturing. We’ll then guide participants through several exercises to identify opportunities, contribute your priorities, comment on the current plans, and learn how to get involved in the future. This is also a great forum to share any concerns about this technology to an audience that can take your feedback into the development plans and have an open discussion about what Autodesk should be doing to address them.

"One of the biggest hurdles in using an STL mesh model for a downward manufacturing application is the lack of geometry feature information in the STL file. This class will demonstrate how machine learning techniques can be applied to extract geometrical information from mesh models; and explain the process of building machine learning models, feature engineering for mesh data, and parameter tuning. We will also cover Autodesk Forge integration for data translation and viewing; and demonstrate how to use AWS cloud infrastructure such as GPU computing for big data and AWS Lambda for fast online predictions."

Machine learning is a much talked about and relatively new way for creating artificial intelligence systems and applying them to complicated problems. KLH Engineers is applying artificial intelligence to solve challenging problems in the architecture, engineering, and construction (AEC) industry. This class is a case study on applying machine learning to the problem of MEP element placement based on BIM (Building Information Modeling) data provided by other consultants. Attendees will discover how KLH applied a machine-learning algorithm to classify BIM objects in a room.

AEC industry is getting heavy datafied and this trend will continue in the next years . Data science enables to extract knowledge of complex data for enhancing decision taking, along all project phases. This class will show how MEP engineers can enhance Dynamotized HVAC analysis by using machine learning techniques. So, the goal is to predict HVAC systems performance, when some of the design parameter inputs are unknown along design process. To be more specific, this class will cope with HVAC design calculations by integrating Autodesk Revit 2018 and Dynamo with Microsoft Power BI using R scripts. The machine learning technique shown will be supervised learning and an example of ventilation and heating systems analysis will be explored. Moreover, it will show best practices to gather BIM historical data and define the explanatory variables set needed to create machine learning algorithms. Machine learning skills are something that everyone can learn and that AEC industry needs.

Integrating building performance analysis results in the early stage of conceptual design supports effective design exploration and decision making. Such integration requires domain expertise and can be technically challenging, time consuming, and computationally expensive. Machine learning demonstrates great potential for building seamless workflows that provide faster analysis feedback. However, one of the challenges in the architecture domain is the lack of diverse data sets to train effective prediction models. This talk will provide a brief overview of the potential of machine learning in building analysis-prediction models and demonstrate how we can capitalize on generative design to compensate for the lack of data. We will present a workflow where generative design features for Revit software, Dynamo software, and Insight software are used to generate a synthetic data set for training a machine-learning model for energy-analysis prediction.

Machine learning is revolutionizing the construction process both on the jobsite and at the desk. Products like Construction IQ, BuildingConnected, Smartvid.io, HoloBuilder, and Versatile help customers predict and prevent risks in real time on construction jobsites and in the office by using data and applying machine learning to it. Forge plays a key role in connecting all these solutions into a seamless network for construction industry. Come and learn from the development leads of these innovative applications about how their applications are using machine learning to improve industry workflows while they discuss best practices, challenges, and future opportunities. You’ll have an opportunity to interact with these leaders and ask your questions.

Machine Learning (ML) techniques are now widely being used in almost all industries. This course focuses on how you can use ML in  the design of Mechanical Systems. We will share our insights on developing a methodology for quickly building a ML design app for the most common control valve type (the Symmetric Butterfly Valve), and how you can use numerical or experimental data generated from simulationHub to train your models. This class will step through machine learning fundamentals, data analysis, training and validating a ML model; and how to generate new 3D designs based on ML using Autodesk Forge.

Project stakeholders need to understand the state of their projects in a simple and easy-to-consume format. Autodesk Construction Cloud users can maximize the power of Construction IQ software to produce predictive insights and display the project analytics in a powerful but easy-to-understand dashboard with live links to the information. Construction IQ, with the help of machine learning, allows the architecture, engineering, and construction (AEC) teams to prioritize issues, and streamline and expedite decision making, thus improving the project delivery outcomes for all stakeholders. This presentation will showcase GHD's use of machine learning within Construction IQ to surface the data on safety and quality, enabling the resolution of issues before they become critical on site. Through this session, you'll better understand how to implement Construction IQ on your project today, gaining insights into projects that otherwise wouldn't be possible.

Machine learning is revolutionizing the construction process both on the job site and at the desk. Products like BIM 360 IQ help customers mitigate risk and avoid blind spots in real time on construction job sites by using data and applying machine learning to it. Our customers use predictive machine-learning components, like subcontractor and project risks, to protect their profit margins from quality issues and stay on track. Come and learn from construction experts and thought leaders from the largest general contractors in the world to find out how these firms are leading the industry by using machine learning to improve company workflows.

For several years, KLH Engineers has led efforts to bridge the gap between AutoCAD and Revit software through machine learning, enhancing our deliverable by building intelligent models from CAD plans. Expanding on our foundational work from 2019, our tool now automatically generates Revit wall elements with precise thickness from AutoCAD drawings, eliminating the problem of thin wall creation and significantly saving time. In this session, we'll demonstrate how machine learning is integral to our process. We have incorporated advanced algorithms that accurately place door and window families within Revit models, using AutoCAD data to inform our workflow and put the "I" back in building information modeling (BIM). Join us to see these advancements in action and learn how they can enhance your design process, making it faster, smarter, and more precise.

The field of construction is well placed to benefit from the advent of machine learning and artificial intelligence. As part of the BIM 360 Project IQ Team at Autodesk, I've had the privilege of being a part of Autodesk's foray into machine learning for construction. This talk will focus on summarizing the developments in this space, and we'll cover some ways in which one can prepare to maximize value from this technology. The class has 2 sections. The first part provides a broad survey of some of the applications of AI and machine learning in construction, and the potential impact. These processes are making changes across various areas, including risk management, schedule management, subcontractor management, construction site environment monitoring, and safety, to name a few. In the second part, the focus will be on construction industry leaders who will talk about their experiences with smarter tools in their daily jobs, and their views of the impact that these tools might have in the short and long terms.

There's a perfect storm of technology change driving the construction revolution today. Mobile devices, Internet of Things, sensors, and drones are continuously capturing construction job-site data and the cloud is aggregating this data. Machine-learning applications such as BIM 360 IQ software help construction project teams analyze risk from this data in real time and take corrective actions. Layton Construction and Autodesk have been collaborating and working together for the past year to apply the BIM 360 IQ machine-learning system for construction safety. In this session, Cooper Darling-an assistant project manager from Layton Construction-will be sharing his project manager experiences involved in helping identify and reduce safety risks on a construction job site with the aid of assistive technologies such as machine learning. Cooper is also part of Layton's National Safety Leadership Team, and he's helping champion the processes and standards based on BIM 360 software for Layton.

"Artificial intelligence (AI) has made a huge impact on making things, bringing with it radical changes in the way things are designed, made, and used. This class will demonstrate how to leverage Amazon Web Services to build, simplify, and streamline intelligent applications for the construction and manufacturing industries. We will walk through machine learning model deployment that can result in 2x better performance on resource constrained platforms such as mobile, desktop and edge devices (IoT). You will walk away from this Forge class with an understanding of how you can use AWS AI/ML services to accelerate innovation and time to market, and to develop new business models."

During our daily office routines, we must accomplish several repetitive and time-consuming tasks to meet client requirements and international standards. Some of these tasks are easy to automate; others seem limited by desktop computing resources and require human intervention. With the evolution of machine learning and introduction of the Forge Design Automation API, EMDC Group was able to automate some of those processes, thus saving tremendous amounts of time and improving our team's efficiency. This class aims to present a sample of what we've built in house to give attendees an idea about the tasks that can nowadays be automated and the potential that artificial intelligence and cloud computing introduce into the architecture, engineering, and construction (AEC) industry.

Sources of video and photo content are rapidly increasing within construction. In addition to mobile phones and tablets, drones and even wearables are starting to generate their own views of “reality” from the job site. How can this information best be capitalized on for value during construction, and managed afterward for future retrieval? In this case study-driven class, Oliver Smith of Skanska and Josh Kanner of Smartvid.io will explore how a new set of technologies are helping to deal with the content proliferation—and are also going beyond to elevate the power and value of this visual data. New techniques in machine learning are enabling mining of the native content to help tag and search the ever-growing set of videos and photos coming from the field. Once tagged, the content can be integrated with Building Information Modeling (BIM) for ongoing progress updates during the project and other workflows. We will present examples of drone-generated and mobile-phone-generated field-progress capture. This session features Navisworks Manage, AutoCAD, and Revit. AIA Approved

Do you have a data explosion? Are you worried about duplicated models and redundant work? Is this a constant battle? Experience ground breaking new technology from Autodesk that uses sophisticated Machine Learning techniques to analyze your 3D design information and provides you with a new way of navigating and discovering patterns in your data. Get an exciting preview of the upcoming Design Graph technology first introduced at AU 2014.

With the Design Graph, 3D content is automatically categorized, correlated, and linked even if there are no explicit relationships between the models and the components within them. In this session you will learn how to explore your design information through these learned categories: shape similarity; relationships between components and designs; parametric properties; and keywords, all at the same time!