The Future of Project Management with Generative Design

From apppm
Jump to: navigation, search

Developed by Jacob Olsen


Figure 1: Generative design illustrated with a bike frame from Autodesk Project Dreamcatcher. [1].

The digital technology has developed exponentially in the past few years and the construction industry has to understand the trends that come with this development as some of the new technologies have come to disrupt the industry, like the case with the transition from CAD systems to BIM systems. Some companies have been living by the principle "disrupt or die" in order to follow the demands and the competition from the market, however overall is the construction industry still lagging behind other industries when it comes to adopting new digital technologies [2]. McKinsey&Company has in a great report recently stated that one of the main subjects to focus on is to infuse digital technology and advanced automation. [3]

Generative design is a technology that has the purpose to imitate the nature's evolutionary approach to design thus it is a radical new way of doing design propositions compared to the traditional way. Shortly described is the generative design concept about expressing the design intent and purpose through e.g. planning strategies, formal strategies or building components, and then get a computer to automatically explore all the options to improve e.g. accuracy of costs and schedule estimates and potential risks thus streamlining the workflow of design. This will enable more structured monitoring of projects and will raise the transparency when the schedule is included in the strategy. This allows the companies to explore a wider range of designs based on criteria and rules and by this rethinking the design and engineering processes with a greater focus on constructability and productivity as an example.

This article will mainly focus on the general aspects of generative design and how it possibly will disrupt the project management and a newly released product from Autodesk will also be shortly introduced. It is important to know that the development of a technology like this is progressing extremely rapidly as generative design is highly prioritised in both software developing companies but also in consulting engineering companies. The potentials and the limitations with the new technology are also discussed with respect to project management.

Contents

Design Technology Trends

Digitisation is a broad term that covers a variety of tools and technologies [4], including BIM (Building Information Modelling (BIM)), automation, simulation and artificial intelligence and these new technology trends open up new ways to work and collaborate. The digitalisation provides more productivity, deliveries of better products and of course a change in how people work. According to Winch is BIM a very important development in IT tools in the history of project management as it places the construction project manager as the key person in the project information flows. [5] In the following sections are BIM, computational design and generative design described shortly.

BIM

For further information, see also BIM-based information flow during construction phase § Application

BIM is not a specific software tool but rather a way of working with 3D models. [6] thus it is a workflow for creating and managing information on a construction project [7].

The difference between the old CAD systems and the new BIM systems is that the software can recognize objects as being walls, columns ect. instead of just lines. Each object has attached intelligent and structured data in a BIM model, thus it is a 3D model with information stored inside the objects and this enables the computers to make calculations on the models. The information in the objects can be utilised to design with computational and generative design i.e. getting more and more automation.

BIM offers extensive opportunities to drive a more effective and efficient built environment. BIM as a project management tool provides the opportunity to achieve more efficient project management processes. If BIM is fully utilised it will improve the planning and delivery of projects thus achieving better outcomes for clients.

CAD stands for computer-aided design but a lot of people with passion for BIM has actually faced that the real description should be computer-aided documentation as the computer doesn't aid the design. The design is in the users head and the computer just document this design by a CAD model. [8] With a BIM model it is actually possible to make the computer aid the design through computational and generative design.

Computational Design

Computational design is basically a design process expressed by using parameters and various rules that together define and clarify the relationship between design intent and design output. For instance structural engineers use finite element analysis (FEA) and MEP engineers use computational fluid dynamics (CFD). With BIM models it is possible to make computations like these directly from the models if the information stored in the objects is sufficient for the analysis.

Table 1: Explaining the B, I and M
B (Building) I (Information) M (Modeling)
  • The act
  • Occupation of constructing
  • Process
  • Datamanagement
  • Consistency
  • Analysis
  • Express and handle complex shapes
  • Autogeneration of geometric shapes
  • Rule based design

Table 1 shows some examples of how each letter in BIM can be interpreted when talking about computational design and generative design.

Generative Design

Figure 2: The concept of computational and generative design [8]

Generative Design is a programming-based approach where design intent and design purpose is expressed. A computer then explores all the options of a solution to find the best solutions. It allows to explore a wider range of design options as in the time a person can create one idea, a computer can generate thousands, along with the data to prove which designs that perform best based on the values and criteria that has been chosen.

Generative Design is about determining the values for a project, and the software will then create high-performing design options based on those criteria thus the software explores conflicting design constraints so the project team can focus on innovation instead of documenting the design with a 3D model. A design can be called generative when it is based on rules and criteria and these set of rules should fulfill the design and the tool should come with multiple design solutions.

A lot of the time spent on projects today is in the detailed design and with generative design it becomes more a competition of how fast a company can do the detailed design rather than a heavy task. If the time used on the detailed design can be half the time used today then the companies will have a huge competitive advantage. Humans will be less in the business of forming a 3D modelling and more in the business of really understanding what the requirements of the design are thus understanding and managing the goals of the project in order to provide solutions that meet the clients and stakeholders expectations. Generative design will lead to better design coordination and this will lead to better project documentation reducing inefficiencies and non-value adding tasks. This will in the end creates better synergies between the project plan, design strategy and the BIM strategy as illustrated in Figure 4.

Figure 3: Generative design illustrated on a floor plan [9]

Application

Autodesk as a frontrunner in the Construction Industry

The founding principal of the Autodesk Studio "The Living", David Benjamin has stated following about generative design:

"Generative design allows us to go beyond some of our typical human linear thinking, our typical rules of thumb – it basically allows us to discover new possibilities" [10]

When it comes to software developing companies then Autodesk is a frontrunner in adapting generative design to the construction industry and Autodesk has recently released the newest BIM software called Project Dreamcatcher. [1]. Project Dreamcatcher is a software that utilises the principle of generative design to enable designers to define their design problem through goals and constraints. These goals and contraints are then used to generate alternative design solutions. An example of a project from Dreamcatcher is showed in Figure 1. This software allows the designers to explore alternative approaches of a possible design and makes it possible for the whole project team to select the design solutions that fit their needs. The workflow should be that the goals and constraints are iterated and adjusted to generate new and better results for each analysis. This process may enable the project team to find interesting solutions that might help to redefine the problem thus repeating the generative loop. This process should make it possible for the designers to generate thousands of solutions to choose from and each solution should be evaluated based on the goals set by the project team as shown in Figure 3.

About 100 years ago in 1915, biologist D’Arcy Wentworth Thompson said,

“The form of an object is a diagram of its forces.”[8].

This statement is the main principle of what Autodesk has adopted in Project Dreamcatcher. The program lets the designers describe the forces that act on an object such as structural loads and then a computer compute several solutions. Generative design will disrupt the market today as it is a paradigm shift that leads to new workflows for people involved in the construction industry.

Generative Design in Project Management

Figure 4: Generative design as part of the project management [11].

Today the construction industry is met with new requirements from the clients and this often happens when the client strives for innovation. With generative design the client, contractor, manufacturer and the stakeholders can be included in an integrated process as illustrated on Figure 4.

In reality, it's not about bits and bytes, but more about getting a better dialogue with the client. With new digital tools it is possible to visualise and simulate solutions so that the client can see the solution before even getting started constructing the building or getting the final design. However digitization is not a solution in itself, but rather a supportive tool that will give clients an experience that they are invited in thus gaining a strong knowledge of many of the things they want, but can't always formulate. [12]

In the future, consultants will spend much more time working with clients on the creative part of the project development. The automation of tasks should not overrule the competences of the engineers or architects in the consulting industry, however the automation should contribute to evolve their competences. Engineers and architects should not worry, although increased automation will remove traditional computing tasks.

How will Generative Design affect the Project Management?

When it comes to project management an important aspect is the implication of generative design on the project manager. The workflow with BIM and project management are both concerning collaboration, coordination, communication, exchange and collation.[13] This means that with more projects and organisations beginning to use BIM in meaningful ways, the role and responsibilities of a project manager are changing.

Generative design may sound simple however one of the challenges is to define rules that make sense resulting in solutions that can be useful. The design process is neither linear or rational in the construction industry as parties tries to work together, where people will sit with each of their speciality. This means that the work flow today is that a design is done, then a better understanding of the problem is achieved and then this process is repeated to iterate to the best solution as illustrated on Figure 2, but this is a hard and very time-consuming task.

The goals, rules and criteria should somehow be defined from a value perspective view. Value is not just cost and it's hard to understand the true meaning of value. Is it even possible to argue that something is more valuable than another thing or should the value be based on a group consent instead. All this depends on the context and the value should be based on valid information in order to be durable. The next thing is what is relevant. Is the client aware of what is possible as the framing and priority of design elements are different from the client to the stakeholder. Generative design reframes the problem and makes it possible to consider other design solutions.

When talking about project management it is often that in the beginning the decisions are very important and later it becomes less important. Generative design lets the client take part in the design thus resulting in a more robust design i.e. it supports decision making in the early design stages. This means that generative design somehow is a part of the value management as it makes it possible to explore priorities early in the project. In the later stages of a project comes the detailed design. The time used on detailed design will become significantly shorter in the future with generative design as it is already defined what to do through the goals and criteria. This means that the first stages in a project will become more creative and here the dialogue with the client and the stakeholders will be significantly bigger.

Managing Disruption

Video 1: Autodesk University Explainers - Managing Disruption with Bill Allen [14]

The construction industry strives for highly efficient workflows in order to create more successful projects. The use of BIM tools and other efficient technologies will give the project manager a strategic role in the paradigm shift as project managers will have to assist the project teams in order to make the implementation successful. The project manager will be the key person to successfully implement the new technologies on projects. With disrupting technologies the market is most likely to experience a paradigm shift. Therefore it is very important that the project manager clearly understands these technologies and paradigms i.e. know how the technologies will support and affect the project management and the project itself.

According to the director of building Information management services at EvolveLAB, Bill Allen, the current processes are going to be changed a lot with the advancements in generative design and software algorithms.[14] A lot of the manually work will in the future be done more and more by computers. Allen thinks that the definition of Building Information Modelling with this advancement of computational and generative design is rather going to be Building Information Optimization as he explains in Video 1. This means that rather than manually drawing e.g. walls and columns, the computer will consider some input rules as instructions to calculate and model the best solution according to e.g. structural load capacity or/and thermal performance. Things that took months before will instead be done in a day. The question is how well the construction industry will implement this new way of working. The project manager has to make an active choice to be a part in this changing process. Project managers need to understand the big picture of the paradigm shift and the implementation of new technologies in order to get the potential to disrupt the industry and have a role in the disruption.

There are two sides of these new disruptive technologies. Some people might say that the creative process and insight is reserved for the engineers and/or designers and this work should a computer not take over. Some might say that the new technologies should be embraced so the computer could make the simple drawing and form creation thus instead focusing on and understanding the real goals and constraints around the problem. A lot of the constraints does not lie as explicit knowledge thus it is important to figure out what the rules should be. However when the rules and goals have been identified, then the solutions should be picked and this can be difficult with thousands of solutions. The computers today have the processor force to find the countless amount of solutions, so it is not the hardware that is the limitation, but rather the companies, people and the whole implementation that is the real limitation of disruptive technologies.

Limitations of disruptive technologies

There's a risk that designers will feel threatened by removing a large part of the creative process from their work and questions will arise around the eventual limits of automation. As a natural result of the automation of tasks, companies will have to implement the new workflows thus provide training and education to prepare the employees for the new workflow. It is important to know that implementation is a huge limitation of new technologies especially when it comes to new technologies that will disrupt the market.

When it comes to new technologies then it is difficult to define what can be done and what can't be done. About 50 years ago it was absurd if someone was asked to know the exact location of 6 billion people, but today everyone walks around with a mobile phone with a GPS so now it's possible to say where the 6 billion people are without asking each and every one. The same principle applies with generative design as with more and more computing power it becomes possible to calculate thousands of possible solutions. The processor power is not the challenge anymore but it is rather a question of how well the construction industry is willing to implement and embrace the new technologies. Current processes will naturally change with advancements in new technologies such as generative design. The next step in the digitalisation is to apply artificial intelligence which will soon be a game changer in the IT tools for design. [15]

References

  1. 1.0 1.1 Autodesk (2017). https://autodeskresearch.com/projects/dreamcatcher. (10-06-2017)
  2. Rob Leslie-Carter, John McGlynn, and Andrew Edkins (2017). Future of project management. Arup and The Bartlett Faculty of the Built Environment at UCL and Association for Project Management 1, 21–24.
  3. McKinsey&Company (2017). Reinventing construction: A route to higher productivity. McKinsey Global Insitute 1, 8–10.
  4. Larsen, R. (2017). Den digitale forandring. http://www.dtu.dk/om-dtu/nyheder-og-presse/aktuelt/2017/06/den-digitale-forandring?id=cc6c4058-7029-4fe8-9788-b05e3470de65. (21-06-2017)
  5. Winch, G. (2010). Managing Construction Projects. Wiley-Blackwell 2, 396.
  6. Koutsogiannis, A. (2017). The Future Of Construction – BIM. https://geniebelt-com.cdn.ampproject.org/c/s/geniebelt.com/blog/bim-maturity-levels/amp. (19-06-2017)
  7. Rothenborg, M. How man and machine can help each other. http://www.ramboll.com/media/rgr/how-man-and-machine-can-help-each-other. (17-06-2017)
  8. 8.0 8.1 8.2 Kowalski, J. (2016). CAD Is a Lie: Generative Design to the Rescue. https://redshift.autodesk.com/generative-design/. (17-06-2017)
  9. Vermeulen, D. (2017). Generative Design Applied on Buildings. http://autodesk.typepad.com/bimtoolbox/2017/06/generative-design-applied-on-buildings.html. (11-06-2017)
  10. Autodesk Toronto (2017). http://autodesktoronto.com/generative-design/. (10-06-2017)
  11. Architectureresearchlab (2012). http://www.architectureresearchlab.com/arl/wp-content/uploads/2012/08/search-construct.jpg. (10-06-2017)
  12. Wessel, L. (2017). Digitalisering giver plads til ingeniørens kreative sider. https://ing-dk.cdn.ampproject.org/c/s/ing.dk/artikel/digitalisering-giverplads-ingeniorens-kreative-sider-200009?amp. (11-06-2017)
  13. Rob Anil Sawhney, Atul R Khanzode, and Saurabh Tiwari (2017). Building Information Modelling for Project Managers. Royal Institution of Chartered Surveyors, 21–24. http://www.rics.org/Global/RICS%20BIM%20for%20Project%20Managers.pdf. (17-06-2017)
  14. 14.0 14.1 Allen, B. (2017). Managing Disruption. https://www.evolvebim.com/single-post/2017/04/13/Managing-Disruption. (17-06-2017)
  15. Higgins, S. (2017). BuildingSP: How to get advanced AI in AEC. http://www.aecst.com/path-advanced-artificial-intelligence-aec/. (19-06-2017)

Annotated Bibliography

Further Reading

Author Title Date Language Category Brief description Link Visited
Jonny Williamson How does generative design unlock engineering innovation? 2017 English Webpage This article explains how far Autodesk vision with generative design has come and the reactions from the engineers and designers. https://www.themanufacturer.com/articles/how-does-generative-design-unlock-engineering-innovation/ 11-06-2017
Bill Allen Implications for Rule-Based Modeling 2017 English Webpage Partner and Director of Building Information Management Services at EvolveLAB Bill Allen discusses the implications for rule-based modeling https://www.evolvebim.com/single-post/2017/04/13/Implications-for-Rule-Based-Modeling 17-06-2017
Rafael Sacks, Paul Teicholz, Chuck Eastman & Kathleen Liston BIM Handbook - A Guide to Building Information Modeling for Owners, Managers, Designers, Engineers and Contractors 2011 English Book The BIM Handbook gives a general and deep understanding of BIM technologies and it points out why it is so difficult to implement new strategies and softwares in the business and organisations. The book also points out the benefits that comes with BIM for all members of a project team.
Astrid Haug Fremtiden tilhører de frygtløse - en guide til den digitale forretningsudvikling 2017 Danish Book The book is mainly about how the companies and organisations can handle the digitalisation and how they can handle the change. The book also concerns subjects like disruption, artificial intelligence and sprints.

Videos

Autodesk presents generative design in a short video.
Autodesk CTO Jeff Kowalski discusses the results driven by generative design in a short video.
Partner and Director of Building Information Management Services at EvolveLAB Bill Allen discusses how the current processes are going to be changed with the advancements in generative design, software algorithms, and robotic construction.
Personal tools
Namespaces

Variants
Actions
Navigation
Toolbox